Search

Inter-project knowledge transfer is by scholars widely considered to be a crucial source of competitiveness for project-based organizations. Effective knowledge transfer not only brings economic benefits but also enhances firms’ environmental and social sustainability performance. However, to date, limited research has focused on ascertaining the prerequisites for effective knowledge transfer within the SME context. Correspondingly, through single-case study, this study aims to understand how interrelated organizational factors impact inter-project knowledge transfer in project-based SMEs. Aspects that were found to impact knowledge transfer could be grouped into three dimensions of organizational factors; project characteristics, organizational structure, and organizational culture. More specifically, project similarities, project complexity, horizontal differentiation, and strong interpersonal relationships were found to facilitate knowledge transfer. On the other hand, urgency, absence of formal project managers, low degree of formalization, centralization, low degree of vertical differentiation, and sense of ownership of knowledge were factors identified to hinder knowledge transfer. Furthermore, the present study discovered how the dimensions of organizational factors are interrelated in their impact on inter-project knowledge transfer. Based on the findings, academic and practical implications, and future research avenues are proposed.

Inter-project knowledge transfer is by scholars widely considered to be a crucial source of competitiveness for project-based organizations. Effective knowledge transfer not only brings economic benefits but also enhances firms’ environmental and social sustainability performance. However, to date, limited research has focused on ascertaining the prerequisites for effective knowledge transfer within the SME context. Correspondingly, through single-case study, this study aims to understand how interrelated organizational factors impact inter-project knowledge transfer in project-based SMEs. Aspects that were found to impact knowledge transfer could be grouped into three dimensions of organizational factors; project characteristics, organizational structure, and organizational culture. More specifically, project similarities, project complexity, horizontal differentiation, and strong interpersonal relationships were found to facilitate knowledge transfer. On the other hand, urgency, absence of formal project managers, low degree of formalization, centralization, low degree of vertical differentiation, and sense of ownership of knowledge were factors identified to hinder knowledge transfer. Furthermore, the present study discovered how the dimensions of organizational factors are interrelated in their impact on inter-project knowledge transfer. Based on the findings, academic and practical implications, and future research avenues are proposed.

Overview: To maximize the rent-yield of the innovation and product development process, firms attempt to commercialize knowledge through activities that require effective knowledge transfer. Knowledge transfer remains a challenge for most firms, however, due to the recipient organization’s lack of ability to absorb and apply new knowledge. We examined four knowledge-transfer projects within a large multinational corporation. We identified important actions, lessons learned, and recommendations that practitioners can use to enhance their own knowledge-transfer processes.

To maximize the rent-yield of the innovation and product development process, firms attempt to commercialize knowledge through activities that require effective knowledge transfer. Knowledge transfer remains a challenge for most firms, however, due to the recipient organization’s lack of ability to absorb and apply new knowledge. We examined four knowledge-transfer projects within a large multinational corporation. We identified important actions, lessons learned, and recommendations that practitioners can use to enhance their own knowledge-transfer processes.

Rapid global change threatens to outstrip global efforts to establish sustainable stewardship of social-ecological systems (SES). Place-based research can enhance effectiveness of global sustainability policies and actions by providing contextualized knowledge underpinning bottom-up solutions. However, the use and transfer of place-based knowledge remains a major challenge. In this study, we analyze place-based knowledge transfer in a local-to-global and knowledge-to-action context. We aim to provide insights on when, how, and why place-based research can inform decision making at the global scale and lead to action toward more sustainable and just futures. Our iterative and exploratory methodology involved alternating rounds of literature reviews and interviews with interdisciplinary researchers. We identified four key steps (place-based knowledge production, knowledge synthesis, knowledge use at the global scale, and knowledge revision and lessons learned) and five facilitative factors (bridging organizations, knowledge brokers, boundary organizations, institutionalized knowledge governance, and polycentric governance systems), which provide a comprehensive understanding of place-based knowledge transfer. Our conceptual framework provides suggestions on how to set up place-based knowledge transfer to be more effective, complete, and inclusive. Furthermore, our study discusses two major structural challenges that currently inhibit place-based knowledge transfer and shows ways forward for science and policy to overcome these. We argue that place-based knowledge transfer can be an effective means to undo dominant power relations and the epistemic status quo and enable a shift from short-termism in science and policy toward more long-term SES goals. Therefore, it is seminal to open up the predominant value system to more diverse knowledge systems, signifying a shift away from global decision making that is guided by neoliberal capitalist principles and over-emphasizes

This thesis investigates the application of knowledge transfer techniques to process event-based data forperson detection in area surveillance. A teacher-student model setup is employed, where both modelsare pretrained on conventional visual data. The teacher model processes visual images to generate targetlabels for the student model trained on event-based data, forming the baseline model. Building onthis, the project incorporates feature-based knowledge transfer, specifically transferring features fromthe Feature Pyramid Network (FPN) component of the Faster R-CNN ResNet-50 FPN network. Resultsindicate that response-based knowledge transfer can effectively finetune models for event-based data.However, feature-based knowledge transfer yields mixed results, requiring more refined techniques forconsistent improvement. The study identifies limitations, including the need for a more diverse dataset,improved preprocessing methods, labeling techniques, and refined feature-based knowledge transfermethods. This research bridges the gap between conventional object detection methods and event-baseddata, enhancing the applicability of event cameras in surveillance applications.

Purpose - With the rise of the gig economy, management positions are increasingly staffed with flexible labor, so-called interim managers. They plunge into organizations for a limited period, operating in a liminal position as partly insider, partly outsider. Although several contributions to their client organizations are acknowledged, it is unknown how the interim manager’s knowledge from previous assignments is made useful in the new context under these particular working conditions. Therefore, the purpose of this paper is to increase the understanding of how the interim manager’s knowledge is transferred to the client organization while operating from a liminal position. Design/methodology/approach - This paper presents an interview-based multiple case study of six interim assignments where knowledge transfer is considered a social and context-dependent process. Findings - The findings unveil the multifaceted nature of the liminal position, which consists of task orientation, time limitation, political detachment and cultural distance. These facets contribute to knowledge transfer in terms of new shared understandings and joint interests, which in turn might create new practices that augment continuous knowledge-sharing patterns. Originality/value - The results contribute to the research on flexible work arrangements by shedding light on how the liminal position, predominantly depicted as an obstacle for the individual, might facilitate knowledge transfer. Through the process of knowledge generation, it is shown how a short-term engagement might enable the organization to increase its knowledge over time.

Tourism industry relies on destination-level knowledge transfer for innovation and competitiveness, yet slow adoption of knowledge management practices hinders the potential benefits. Academic studies on knowledge transfer and its efficiency are limited and very few have addressed this phenomenon at inter-organizational level particularly in a tourism destination context. We aim to address this gap by providing a conceptual model to extend research knowledge in this area. Drawing on social capital theory and knowledge management constructs, the model incorporates four pivotal dimensions: ‘structural,’ ‘relational,’ ‘organizational,’ and ‘knowledge’ properties, identified as effective antecedents of knowledge transfer through prior research. The proposed model is designed to be practically applicable and measurable in a tourism destination and includes a measurement approach based on a network perspective. The benefits of using such a model include a comparison between destinations, and the opportunity to illuminate the restraints within a specific destination for subsequent strategic management action.

This article examines the central concept of knowledge transfer within organizations, highlighting its decisive role in promoting innovation and organizational learning. Building on the research work of Nonaka and Takeuchi (1995), which analyzes the complex mechanisms of knowledge transfer, including internalization, socialization, externalization and combination, demonstrating that this process goes beyond the simple transfer of information to result in the transformation of individual knowledge into collective and organizational knowledge. The article adopts a conceptual review approach, aimed at structuring and deepening understanding of the complex dynamics inherent in knowledge transfer. This methodological choice makes it possible to propose an integrated theoretical framework, shedding light on the underlying mechanisms and suggesting perspectives for future research. In addition, the article explores the positive Influence of integrating information technologies into the knowledge transfer process, highlighting the advantages offered by knowledge management platforms and enterprise collaborative tools for storing, organizing and disseminating knowledge.  This article also highlights the strategic issues inherent in knowledge management in an organizational context characterized by perpetual dynamics. The in-depth analysis reveals the imperative need to develop methodologically rigorous and contextually adapted practices to optimize the effectiveness of knowledge transfer processes, while integrating the cultural, structural and institutional specificities peculiar to each organization. However, it also recognizes the challenges associated with the adoption of these technologies, such as their user-friendliness and their ability to encourage the commitment of employee users. This article offers a global perspective on knowledge transfer, highlighting its crucial importance for contemporary organizations, while calling for a proactive approach to overcomi, This article examines the central concept of knowledge transfer within organizations, highlighting its decisive role in promoting innovation and organizational learning. Building on the research work of Nonaka and Takeuchi (1995), which analyzes the complex mechanisms of knowledge transfer, including internalization, socialization, externalization and combination, demonstrating that this process goes beyond the simple transfer of information to result in the transformation of individual knowledge into collective and organizational knowledge. The article adopts a conceptual review approach, aimed at structuring and deepening understanding of the complex dynamics inherent in knowledge transfer. This methodological choice makes it possible to propose an integrated theoretical framework, shedding light on the underlying mechanisms and suggesting perspectives for future research. In addition, the article explores the positive Influence of integrating information technologies into the knowledge transfer process, highlighting the advantages offered by knowledge management platforms and enterprise collaborative tools for storing, organizing and disseminating knowledge.  This article also highlights the strategic issues inherent in knowledge management in an organizational context characterized by perpetual dynamics. The in-depth analysis reveals the imperative need to develop methodologically rigorous and contextually adapted practices to optimize the effectiveness of knowledge transfer processes, while integrating the cultural, structural and institutional specificities peculiar to each organization. However, it also recognizes the challenges associated with the adoption of these technologies, such as their user-friendliness and their ability to encourage the commitment of employee users. This article offers a global perspective on knowledge transfer, highlighting its crucial importance for contemporary organizations, while calling for a proactive approach to overcomi

This thesis investigates the dynamics of knowledge transfer between cultures within multinational firms, with a particular emphasis on subsidiaries located in Sweden and Romania. This study investigates how subsidiaries adjust corporate knowledge to fit local cultural contexts through semi-structured interviews with participants from a range of sectors. It also looks at the wider effects of these changes on organizational knowledge transfer processes. The research methodology used is qualitative research. The data indicates that local cultural and regulatory variations have an impact on knowledge transfer tactics. Different techniques to integrating and adjusting corporate knowledge are shown by companies in Sweden and Romania, which reflect different operational and cultural contexts. The results show how subsidiaries improve their strategic positioning and operational efficacy within the multinational framework by utilizing both local and global advantages. The study's methodology involves using thematic analysis to analyze the gathered interview data in order to provide a deeper understanding of the mechanisms underlying cross-cultural knowledge transfer. The main results suggest that the effectiveness of knowledge transfer is heavily dependent upon the ability to navigate and incorporate local cultural characteristics alongside international standards. By offering empirical insights into the difficulties and methods of knowledge transfer within Eastern European contexts, which are less studied than those in the West, this study adds to the body of literature already in existence. Additionally, it has useful ramifications for global firms looking to maximize knowledge management in various cultural contexts.

This thesis investigates the application of knowledge transfer techniques to process event-based data forperson detection in area surveillance. A teacher-student model setup is employed, where both modelsare pretrained on conventional visual data. The teacher model processes visual images to generate targetlabels for the student model trained on event-based data, forming the baseline model. Building onthis, the project incorporates feature-based knowledge transfer, specifically transferring features fromthe Feature Pyramid Network (FPN) component of the Faster R-CNN ResNet-50 FPN network. Resultsindicate that response-based knowledge transfer can effectively finetune models for event-based data.However, feature-based knowledge transfer yields mixed results, requiring more refined techniques forconsistent improvement. The study identifies limitations, including the need for a more diverse dataset,improved preprocessing methods, labeling techniques, and refined feature-based knowledge transfermethods. This research bridges the gap between conventional object detection methods and event-baseddata, enhancing the applicability of event cameras in surveillance applications.

To harness the knowledge, skill and ability of the workforce demands that companies are intentional about human resource practices and processes, not only to retain workers but also equip them to maximise employee output. Critical among these practices is Onboarding. This research investigates how knowledge is transferred during onboarding in organisations in the post covid era, focusing on organisations that heavily rely on employee knowledge to gain a competitive advantage. For such organisations, business success depends on the unique knowledge and skill sets of their employees. The study employs a qualitative research strategy with a case study method to enable an in-depth examination of the subject in an organisation. Data were collected through semi-structured interviews with a purposive sampling of employees, managers, and HR professionals, as well as through organisational materials. The interviews were conducted remotely via Microsoft Teams, with measures taken to ensure data quality and adhere to ethical principles, including participant anonymity and consent. The findings highlight the need to reorganize company induction manuals, standardize procedures for role-specific onboarding, and establish mentorship programs to improve knowledge transfer during onboarding after covid. These changes can enhance organisational knowledge transfer processes, leading to more effective operations and improved onboarding experiences for new employees in the post-Covid era.

Knowledge transfer encompasses not only scientific communication, but all forms of knowledge exchange between science and various social groups. To quantify the potential for knowledge transfer, the Berlin Science Survey looked at the perspectives of researchers and asked them for which areas of society they consider their own research to be relevant. The transfer potential is to a large extent already realized. Of all respondents, 73% are in contact with at least one social group; 54% are even in contact with two or more groups. However, different disciplines have very different profiles and therefore different transfer potentials.

This thesis investigates the dynamics of knowledge transfer between cultures within multinational firms, with a particular emphasis on subsidiaries located in Sweden and Romania. This study investigates how subsidiaries adjust corporate knowledge to fit local cultural contexts through semi-structured interviews with participants from a range of sectors. It also looks at the wider effects of these changes on organizational knowledge transfer processes. The research methodology used is qualitative research. The data indicates that local cultural and regulatory variations have an impact on knowledge transfer tactics. Different techniques to integrating and adjusting corporate knowledge are shown by companies in Sweden and Romania, which reflect different operational and cultural contexts. The results show how subsidiaries improve their strategic positioning and operational efficacy within the multinational framework by utilizing both local and global advantages. The study's methodology involves using thematic analysis to analyze the gathered interview data in order to provide a deeper understanding of the mechanisms underlying cross-cultural knowledge transfer. The main results suggest that the effectiveness of knowledge transfer is heavily dependent upon the ability to navigate and incorporate local cultural characteristics alongside international standards. By offering empirical insights into the difficulties and methods of knowledge transfer within Eastern European contexts, which are less studied than those in the West, this study adds to the body of literature already in existence. Additionally, it has useful ramifications for global firms looking to maximize knowledge management in various cultural contexts.

To harness the knowledge, skill and ability of the workforce demands that companies are intentional about human resource practices and processes, not only to retain workers but also equip them to maximise employee output. Critical among these practices is Onboarding. This research investigates how knowledge is transferred during onboarding in organisations in the post covid era, focusing on organisations that heavily rely on employee knowledge to gain a competitive advantage. For such organisations, business success depends on the unique knowledge and skill sets of their employees. The study employs a qualitative research strategy with a case study method to enable an in-depth examination of the subject in an organisation. Data were collected through semi-structured interviews with a purposive sampling of employees, managers, and HR professionals, as well as through organisational materials. The interviews were conducted remotely via Microsoft Teams, with measures taken to ensure data quality and adhere to ethical principles, including participant anonymity and consent. The findings highlight the need to reorganize company induction manuals, standardize procedures for role-specific onboarding, and establish mentorship programs to improve knowledge transfer during onboarding after covid. These changes can enhance organisational knowledge transfer processes, leading to more effective operations and improved onboarding experiences for new employees in the post-Covid era.

It is well known that Chinese multinationals’ international expansion is often supported by the home-country government. This study argues that the political embeddedness of Chinese multinationals’ headquarters may impair their managerial intentions and abilities to carry out their entrepreneurial and administrative roles and, indirectly, their capacity to achieve reverse knowledge transfers from their foreign subsidiaries. Based on a study of 177 subsidiaries of 99 Chinese multinationals, we find no support for the argument that political embeddedness affects headquarters’ centralization of subsidiary decisions, although such centralization is negatively associated with reverse knowledge transfer. Additionally, our findings show that home-country political embeddedness is negatively associated with headquarters’ entrepreneurial role vis-à-vis individual subsidiaries. This hampers subsidiaries’ reverse knowledge transfers. The results contribute to understanding the value of political embeddedness and headquarters’ roles in the context of Chinese multinationals. They also provide insights into the strategic asset-seeking internationalization of Chinese multinationals. By developing and operationalizing the multidimensional concept of political embeddedness, this study further contributes to the research on firm-government relations.

Despite the recognized importance of subsidiary reverse knowledge transfer (RKT) for multinational enterprises' (MNEs') competitive advantages, little is understood about how host country policy uncertainty may affect it. This study, based on absorptive capacity and resource dependence theories and utilizing the global economic policy uncertainty (GEPU) index, examines 1565 firm-year observations of Taiwanese MNE parent–subsidiary activities. Findings suggest that GEPU indirectly undermines subsidiary RKT benefit by reducing MNEs' global R&D intensity and geographical R&D diversity. Notably, this effect is mitigated when MNEs choose joint venture entry modes. This study enriches international business literature by elucidating the intricate relationships between the host country policy uncertainty, MNE strategies, and subsidiary RKT outcomes, thus contributing to a deeper scholarly understanding in the RKT domain. © 2024 European Academy of Management (EURAM).

This article studies the problem of semi-supervised learning on graphs, which aims to incorporate ubiquitous unlabeled knowledge (e.g., graph topology, node attributes) with few-available labeled knowledge (e.g., node class) to alleviate the scarcity issue of supervised information on node classification. While promising results are achieved, existing works for this problem usually suffer from the poor balance of generalization and fitting ability due to the heavy reliance on labels or task-agnostic unsupervised information. To address the challenge, we propose a dual-channel framework for semi-supervised learning on Graphs via Knowledge Transfer between independent supervised and unsupervised embedding spaces, namely, GKT. Specifically, we devise a dual-channel framework including a supervised model for learning the label probability of nodes and an unsupervised model for extracting information from massive unlabeled graph data. A knowledge transfer head is proposed to bridge the gap between the generalization and fitting capability of the two models. We use the unsupervised information to reconstruct batch-graphs to smooth the label probability distribution on the graphs to improve the generalization of prediction. We also adaptively adjust the reconstructed graphs by encouraging the label-related connections to solidify the fitting ability. Since the optimization of the supervised channel with knowledge transfer contains that of the unsupervised channel as a constraint and vice versa, we then propose a meta-learning-based method to solve the bi-level optimization problem, which avoids the negative transfer and further improves the model’s performance. Finally, extensive experiments validate the effectiveness of our proposed framework by comparing state-of-the-art algorithms. CCS Concepts: • Theory of computation → Semi-supervised learning; • Computing methodologies → Unsupervised learning; Transfer learning © 2023 Copyright held by the owner/author(s). Publicatio

It is well known that Chinese multinationals’ international expansion is often supported by the home-country government. This study argues that the political embeddedness of Chinese multinationals’ headquarters may impair their managerial intentions and abilities to carry out their entrepreneurial and administrative roles and, indirectly, their capacity to achieve reverse knowledge transfers from their foreign subsidiaries. Based on a study of 177 subsidiaries of 99 Chinese multinationals, we find no support for the argument that political embeddedness affects headquarters’ centralization of subsidiary decisions, although such centralization is negatively associated with reverse knowledge transfer. Additionally, our findings show that home-country political embeddedness is negatively associated with headquarters’ entrepreneurial role vis-à-vis individual subsidiaries. This hampers subsidiaries’ reverse knowledge transfers. The results contribute to understanding the value of political embeddedness and headquarters’ roles in the context of Chinese multinationals. They also provide insights into the strategic asset-seeking internationalization of Chinese multinationals. By developing and operationalizing the multidimensional concept of political embeddedness, this study further contributes to the research on firm-government relations.

It is well known that Chinese multinationals’ international expansion is often supported by the home-country government. This study argues that the political embeddedness of Chinese multinationals’ headquarters may impair their managerial intentions and abilities to carry out their entrepreneurial and administrative roles and, indirectly, their capacity to achieve reverse knowledge transfers from their foreign subsidiaries. Based on a study of 177 subsidiaries of 99 Chinese multinationals, we find no support for the argument that political embeddedness affects headquarters’ centralization of subsidiary decisions, although such centralization is negatively associated with reverse knowledge transfer. Additionally, our findings show that home-country political embeddedness is negatively associated with headquarters’ entrepreneurial role vis-à-vis individual subsidiaries. This hampers subsidiaries’ reverse knowledge transfers. The results contribute to understanding the value of political embeddedness and headquarters’ roles in the context of Chinese multinationals. They also provide insights into the strategic asset-seeking internationalization of Chinese multinationals. By developing and operationalizing the multidimensional concept of political embeddedness, this study further contributes to the research on firm-government relations.

It is well known that Chinese multinationals’ international expansion is often supported by the home-country government. This study argues that the political embeddedness of Chinese multinationals’ headquarters may impair their managerial intentions and abilities to carry out their entrepreneurial and administrative roles and, indirectly, their capacity to achieve reverse knowledge transfers from their foreign subsidiaries. Based on a study of 177 subsidiaries of 99 Chinese multinationals, we find no support for the argument that political embeddedness affects headquarters’ centralization of subsidiary decisions, although such centralization is negatively associated with reverse knowledge transfer. Additionally, our findings show that home-country political embeddedness is negatively associated with headquarters’ entrepreneurial role vis-à-vis individual subsidiaries. This hampers subsidiaries’ reverse knowledge transfers. The results contribute to understanding the value of political embeddedness and headquarters’ roles in the context of Chinese multinationals. They also provide insights into the strategic asset-seeking internationalization of Chinese multinationals. By developing and operationalizing the multidimensional concept of political embeddedness, this study further contributes to the research on firm-government relations.

This dissertation explores how social interactions within specific boundary organizations (BOs) included in universities’ Third Mission: Knowledge Transfer Offices (KTOs) impact the knowledge shared with society. While new activities and unseen actors continuously shape how universities contribute to society, theorists continue to collect missing pieces to the puzzle of universities’ impact on society, neglecting the relational aspect. This dissertation adopts a relational approach to knowledge transfer and examines three projects undertaken by the KTO of a top-ranked European university. The first longitudinal case study takes a longer-term view. It offers a contextualized analysis of the interactions that shaped the creation and implementation of the university makerspace, run by the KTO. It shows how materiality hinders and facilitates decision-making that potentially impacts the innovation developed inside that space. The second exploratory case study of a digitalized hackathon adopts a micro approach following the creation and development of nine digital objects actors use to engage and sustain collaboration during a hackathon. This study shows how the objects acted as a system and how their form, content, and medium changed through the process and according to the different functions they pursued. It portrays a dynamic view of material objects as carriers of knowledge and situational traces that ultimately shaped the innovative output of the project. The third case study takes a fine-grained approach, combining video-ethnography, conversation analysis, and micro-sociology to trace the second-by-second interactions unfolding in a meeting where a project takes an unexpected turn. The findings highlight a micro-interactional process of creative re-framing that we label ‘emotional scaffolding’ by which team members build a positive emotional mood while connecting existing and new ideas into a new solution. Finally, this thesis discusses the contributions to innova

Conflict Related Sexual Violence (CRSV) is a persistent and pervasive humanitarian issue, exacerbated by complexities inherent in global conflicts. Recognising the widespread character of CRSV, the need for comprehensive measures to combat it becomes apparent. In the Democratic Republic of Congo (DRC), where CRSV remains a prevalent element in the conflict, Panzi Hospital stands out for its specialisation in assisting survivors. The hospital has developed a holistic care model known as the Panzi Model, which serves as a guiding framework for their efforts. In an increasingly interconnected world, characterised by blurred borders and interdependence among societies, the importance of knowledge transfer (KT), especially while acknowledging the importance of an intersectional approach, becomes evident. This study aimed to assess stakeholders' perceptions of the transferability of the Panzi Model, with a specific focus on its transfer from the DRC to Ukraine, facilitated by the Mukwege Foundation. Employing a qualitative mixed-methods approach, empirical data was collected through interviews and document analysis. The analysis was guided by a theoretical framework—a process model for KT developed within the scope of this research, integrating intersectionality into all its phases. Key findings reveal optimism regarding the model's adaptability, but significant challenges persist, including cultural barriers and a lack of conscious decision-making processes of KT. While efforts have been made to raise awareness and provide support services in Ukraine, challenges remain, including a disproportionate focus on advocacy efforts, leaving limited space for tangible medical and psychological support efforts––both key components of the Panzi Model. Findings also revealed the complexities of deciding whether to establish one-stop-centres (OSCs) or to work through referral systems. In conclusion, the study suggests that while the philosophy of the Panzi Model can be transferred, i

Conflict Related Sexual Violence (CRSV) is a persistent and pervasive humanitarian issue, exacerbated by complexities inherent in global conflicts. Recognising the widespread character of CRSV, the need for comprehensive measures to combat it becomes apparent. In the Democratic Republic of Congo (DRC), where CRSV remains a prevalent element in the conflict, Panzi Hospital stands out for its specialisation in assisting survivors. The hospital has developed a holistic care model known as the Panzi Model, which serves as a guiding framework for their efforts. In an increasingly interconnected world, characterised by blurred borders and interdependence among societies, the importance of knowledge transfer (KT), especially while acknowledging the importance of an intersectional approach, becomes evident. This study aimed to assess stakeholders' perceptions of the transferability of the Panzi Model, with a specific focus on its transfer from the DRC to Ukraine, facilitated by the Mukwege Foundation. Employing a qualitative mixed-methods approach, empirical data was collected through interviews and document analysis. The analysis was guided by a theoretical framework—a process model for KT developed within the scope of this research, integrating intersectionality into all its phases. Key findings reveal optimism regarding the model's adaptability, but significant challenges persist, including cultural barriers and a lack of conscious decision-making processes of KT. While efforts have been made to raise awareness and provide support services in Ukraine, challenges remain, including a disproportionate focus on advocacy efforts, leaving limited space for tangible medical and psychological support efforts––both key components of the Panzi Model. Findings also revealed the complexities of deciding whether to establish one-stop-centres (OSCs) or to work through referral systems. In conclusion, the study suggests that while the philosophy of the Panzi Model can be transferred, i

Background: Forest fires represent a severe threat to Mediterranean ecosystems and are considered one of the major environmental and socioeconomic problems of the region. The project Plantando cara al fuego (PCF, Spain) is designed to transfer knowledge and to improve the training of new generations in forest fire management. The project is based on the application of service-learning (S-L), an educational methodology that combines learning and community service. Conceived as a compendium of S-L initiatives, the PCF project is composed of several S-L projects with the objective of reducing the problem of forest fires. The individual projects are developed at the regional/local level, each one involving different social agents such as researchers, students (from different disciplines), schools, multidisciplinary professionals, NGOs, or the administration. Results: Participants received an initial training in S-L to design projects focused on different aspects of forest fires (environmental awareness, outreach/communication, fire prevention or post-fire restoration). These applied projects are formally integrated in the learning process via curriculum, which serve to teach and reinforce transversal skills and allow students to get involved and work to solve real problems. In general, the response of the participants was highly favorable, since the projects served to create an atmosphere that facilitates learning, interaction between participants, the application of theoretical class contents, knowledge transfer, or the exchange of good teaching practices. Conclusions: The summary of the PCF project presented in this work serves as a practical guide describing the activities, participants, and the necessary steps involved in the design, development, and evaluation of S-L projects to address environmental problems. In this case, the S-L was adapted to a specific context (i.e., the problem of forest fires) to which it had never been previously applied, but this methodolo

This study adopts a temporal perspective to investigate how boundary spanners can increase the inflow of external knowledge by engaging with both external and internal parties. We add to prior work on knowledge transfer by shifting the focus from engagement levels to investigating engagement dynamics, especially the degree of switching between external and internal engagement across consecutive time periods. Drawing from a cognitive perspective, we argue that switching strongly between engagement types is associated with a segmented knowledge structure that enables quick and efficient categorical processing when knowledge can simply be "channeled"from source to recipient units. In contrast, weak or no switching is associated with a blended knowledge structure and more reflective processing, which is particularly helpful when knowledge transfer requires more translation and transformation. Correspondingly, we adopt a contingency perspective and theorize that the cognitive advantages associated with stronger versus weaker switching weigh differently, contingent on the stickiness of knowledge to be transferred and the nature of boundary-spanning activities that vary in importance over time. Fixed effects models of eight waves of original survey data reveal that, in line with our theorizing, the association between switching and knowledge transfer becomes increasingly negative (1) the more boundary spanners access knowledge that is transspecialist in nature, (2) the greater the organizational distance.

The aim of this research is to refine knowledge transfer on audio-image temporal agreement for audio-text cross retrieval. To address the limited availability of paired non-speech audio-text data, learning methods for transferring the knowledge acquired from a large amount of paired audio-image data to shared audio-text representation have been investigated, suggesting the importance of how audio-image co-occurrence is learned. Conventional approaches in audio-image learning assign a single image randomly selected from the corresponding video stream to the entire audio clip, assuming their co-occurrence. However, this method may not accurately capture the temporal agreement between the target audio and image because a single image can only represent a snapshot of a scene, though the target audio changes from moment to moment. To address this problem, we propose two methods for audio and image matching that effectively capture the temporal information: (i) Nearest Match wherein an image is selected from multiple time frames based on similarity with audio, and (ii) Multiframe Match wherein audio and image pairs of multiple time frames are used. Experimental results show that method (i) improves the audio-text retrieval performance by selecting the nearest image that aligns with the audio information and transferring the learned knowledge. Conversely, method (ii) improves the performance of audio-image retrieval while not showing significant improvements in audio-text retrieval performance. These results indicate that refining audio-image temporal agreement may contribute to better knowledge transfer to audio-text retrieval., Comment: Submitted to EUSIPCO2024

Transfer learning has aroused great interest in the statistical community. In this article, we focus on knowledge transfer for unsupervised learning tasks in contrast to the supervised learning tasks in the literature. Given the transferable source populations, we propose a two-step transfer learning algorithm to extract useful information from multiple source principal component analysis (PCA) studies, thereby enhancing estimation accuracy for the target PCA task. In the first step, we integrate the shared subspace information across multiple studies by a proposed method named as Grassmannian barycenter, instead of directly performing PCA on the pooled dataset. The proposed Grassmannian barycenter method enjoys robustness and computational advantages in more general cases. Then the resulting estimator for the shared subspace from the first step is further utilized to estimate the target private subspace in the second step. Our theoretical analysis credits the gain of knowledge transfer between PCA studies to the enlarged eigenvalue gap, which is different from the existing supervised transfer learning tasks where sparsity plays the central role. In addition, we prove that the bilinear forms of the empirical spectral projectors have asymptotic normality under weaker eigenvalue gap conditions after knowledge transfer. When the set of informativesources is unknown, we endow our algorithm with the capability of useful dataset selection by solving a rectified optimization problem on the Grassmann manifold, which in turn leads to a computationally friendly rectified Grassmannian K-means procedure. In the end, extensive numerical simulation results and a real data case concerning activity recognition are reported to support our theoretical claims and to illustrate the empirical usefulness of the proposed transfer learning methods.

In recent years, the recommendation content on e-commerce platforms has become increasingly rich -- a single user feed may contain multiple entities, such as selling products, short videos, and content posts. To deal with the multi-entity recommendation problem, an intuitive solution is to adopt the shared-network-based architecture for joint training. The idea is to transfer the extracted knowledge from one type of entity (source entity) to another (target entity). However, different from the conventional same-entity cross-domain recommendation, multi-entity knowledge transfer encounters several important issues: (1) data distributions of the source entity and target entity are naturally different, making the shared-network-based joint training susceptible to the negative transfer issue, (2) more importantly, the corresponding feature schema of each entity is not exactly aligned (e.g., price is an essential feature for selling product while missing for content posts), making the existing methods no longer appropriate. Recent researchers have also experimented with the pre-training and fine-tuning paradigm. Again, they only consider the scenarios with the same entity type and feature systems, which is inappropriate in our case. To this end, we design a pre-training & fine-tuning based Multi-entity Knowledge Transfer framework called MKT. MKT utilizes a multi-entity pre-training module to extract transferable knowledge across different entities. In particular, a feature alignment module is first applied to scale and align different feature schemas. Afterward, a couple of knowledge extractors are employed to extract the common and entity-specific knowledge. In the end, the extracted common knowledge is adopted for target entity model training. Through extensive offline and online experiments, we demonstrated the superiority of MKT over multiple State-Of-The-Art methods.

Chinese Spelling Correction (CSC) aims to detect and correct spelling errors in given sentences. Recently, multi-domain CSC has gradually attracted the attention of researchers because it is more practicable. In this paper, we focus on the key flaw of the CSC model when adapting to multi-domain scenarios: the tendency to forget previously acquired knowledge upon learning new domain-specific knowledge (i.e., catastrophic forgetting). To address this, we propose a novel model-agnostic Multi-stage Knowledge Transfer (MKT) framework, which utilizes a continuously evolving teacher model for knowledge transfer in each domain, rather than focusing solely on new domain knowledge. It deserves to be mentioned that we are the first to apply continual learning methods to the multi-domain CSC task. Experiments prove the effectiveness of our proposed method, and further analyses demonstrate the importance of overcoming catastrophic forgetting for improving the model performance.

In this paper, we propose a novel knowledge transfer framework that introduces continuous normalizing flows for progressive knowledge transformation and leverages multi-step sampling strategies to achieve precision knowledge transfer. We name this framework Knowledge Transfer with Flow Matching (FM-KT), which can be integrated with a metric-based distillation method with any form (\textit{e.g.} vanilla KD, DKD, PKD and DIST) and a meta-encoder with any available architecture (\textit{e.g.} CNN, MLP and Transformer). By introducing stochastic interpolants, FM-KD is readily amenable to arbitrary noise schedules (\textit{e.g.}, VP-ODE, VE-ODE, Rectified flow) for normalized flow path estimation. We theoretically demonstrate that the training objective of FM-KT is equivalent to minimizing the upper bound of the teacher feature map or logit negative log-likelihood. Besides, FM-KT can be viewed as a unique implicit ensemble method that leads to performance gains. By slightly modifying the FM-KT framework, FM-KT can also be transformed into an online distillation framework OFM-KT with desirable performance gains. Through extensive experiments on CIFAR-100, ImageNet-1k, and MS-COCO datasets, we empirically validate the scalability and state-of-the-art performance of our proposed methods among relevant comparison approaches.

Recent advances in training multilingual language models on large datasets seem to have shown promising results in knowledge transfer across languages and achieve high performance on downstream tasks. However, we question to what extent the current evaluation benchmarks and setups accurately measure zero-shot cross-lingual knowledge transfer. In this work, we challenge the assumption that high zero-shot performance on target tasks reflects high cross-lingual ability by introducing more challenging setups involving instances with multiple languages. Through extensive experiments and analysis, we show that the observed high performance of multilingual models can be largely attributed to factors not requiring the transfer of actual linguistic knowledge, such as task- and surface-level knowledge. More specifically, we observe what has been transferred across languages is mostly data artifacts and biases, especially for low-resource languages. Our findings highlight the overlooked drawbacks of existing cross-lingual test data and evaluation setups, calling for a more nuanced understanding of the cross-lingual capabilities of multilingual models., Comment: Accepted to EACL 2024

Source-free cross-modal knowledge transfer is a crucial yet challenging task, which aims to transfer knowledge from one source modality (e.g., RGB) to the target modality (e.g., depth or infrared) with no access to the task-relevant (TR) source data due to memory and privacy concerns. A recent attempt leverages the paired task-irrelevant (TI) data and directly matches the features from them to eliminate the modality gap. However, it ignores a pivotal clue that the paired TI data could be utilized to effectively estimate the source data distribution and better facilitate knowledge transfer to the target modality. To this end, we propose a novel yet concise framework to unlock the potential of paired TI data for enhancing source-free cross-modal knowledge transfer. Our work is buttressed by two key technical components. Firstly, to better estimate the source data distribution, we introduce a Task-irrelevant data-Guided Modality Bridging (TGMB) module. It translates the target modality data (e.g., infrared) into the source-like RGB images based on paired TI data and the guidance of the available source model to alleviate two key gaps: 1) inter-modality gap between the paired TI data; 2) intra-modality gap between TI and TR target data. We then propose a Task-irrelevant data-Guided Knowledge Transfer (TGKT) module that transfers knowledge from the source model to the target model by leveraging the paired TI data. Notably, due to the unavailability of labels for the TR target data and its less reliable prediction from the source model, our TGKT model incorporates a self-supervised pseudo-labeling approach to enable the target model to learn from its predictions. Extensive experiments show that our method achieves state-of-the-art performance on three datasets (RGB-to-depth and RGB-to-infrared).

The advancement of science relies on the exchange of ideas across disciplines and the integration of diverse knowledge domains. However, tracking knowledge flows and interdisciplinary integration in rapidly evolving, multidisciplinary fields remains a significant challenge. This work introduces a novel network analysis framework to study the dynamics of knowledge transfer directly from citation data. By applying dynamic community detection to cumulative, time-evolving citation networks, we can identify research areas as groups of papers sharing knowledge sources and outputs. Our analysis characterises the life-cycles and knowledge transfer patterns of these dynamic communities over time. We demonstrate our approach through a case study of eXplainable Artificial Intelligence (XAI) research, an emerging interdisciplinary field at the intersection of machine learning, statistics, and psychology. Key findings include: (i) knowledge transfer between these important foundational topics and the contemporary topics in XAI research is limited, and the extent of knowledge transfer varies across different contemporary research topics; (ii) certain application domains exist as isolated "knowledge silos"; (iii) significant "knowledge gaps" are identified between related XAI research areas, suggesting opportunities for cross-pollination and improved knowledge integration. By mapping interdisciplinary integration and bridging knowledge gaps, this work can inform strategies to synthesise ideas from disparate sources and drive innovation. More broadly, our proposed framework enables new insights into the evolution of knowledge ecosystems directly from citation data, with applications spanning literature review, research planning, and science policy.

Recent research in federated large language models (LLMs) has primarily focused on enabling clients to fine-tune their locally deployed homogeneous LLMs collaboratively or on transferring knowledge from server-based LLMs to small language models (SLMs) at downstream clients. However, a significant gap remains in the simultaneous mutual enhancement of both the server's LLM and clients' SLMs. To bridge this gap, we propose FedMKT, a parameter-efficient federated mutual knowledge transfer framework for large and small language models. This framework is designed to adaptively transfer knowledge from the server's LLM to clients' SLMs while concurrently enriching the LLM with clients' unique domain insights. We facilitate token alignment using minimum edit distance (MinED) and then selective mutual knowledge transfer between client-side SLMs and a server-side LLM, aiming to collectively enhance their performance. Through extensive experiments across three distinct scenarios, heterogeneous, homogeneous, and one-to-one, we evaluate the effectiveness of FedMKT using various public LLMs and SLMs on a range of NLP text generation tasks. Empirical results demonstrate significant performance improvements in clients' SLMs with the aid of the LLM. Furthermore, the LLM optimized by FedMKT achieves a performance comparable to that achieved through direct fine-tuning based on clients' data, highlighting the effectiveness and adaptability of FedMKT.

This paper investigates the accuracy of generative models and the impact of knowledge transfer on their generation precision. Specifically, we examine a generative model for a target task, fine-tuned using a pre-trained model from a source task. Building on the "Shared Embedding" concept, which bridges the source and target tasks, we introduce a novel framework for transfer learning under distribution metrics such as the Kullback-Leibler divergence. This framework underscores the importance of leveraging inherent similarities between diverse tasks despite their distinct data distributions. Our theory suggests that the shared structures can augment the generation accuracy for a target task, reliant on the capability of a source model to identify shared structures and effective knowledge transfer from source to target learning. To demonstrate the practical utility of this framework, we explore the theoretical implications for two specific generative models: diffusion and normalizing flows. The results show enhanced performance in both models over their non-transfer counterparts, indicating advancements for diffusion models and providing fresh insights into normalizing flows in transfer and non-transfer settings. These results highlight the significant contribution of knowledge transfer in boosting the generation capabilities of these models.

As large language models (LLMs) demonstrate unparalleled performance and generalization ability, LLMs are widely used and integrated into various applications. When it comes to sensitive domains, as commonly described in federated learning scenarios, directly using external LLMs on private data is strictly prohibited by stringent data security and privacy regulations. For local clients, the utilization of LLMs to improve the domain-specific small language models (SLMs), characterized by limited computational resources and domain-specific data, has attracted considerable research attention. By observing that LLMs can empower domain-specific SLMs, existing methods predominantly concentrate on leveraging the public data or LLMs to generate more data to transfer knowledge from LLMs to SLMs. However, due to the discrepancies between LLMs' generated data and clients' domain-specific data, these methods cannot yield substantial improvements in the domain-specific tasks. In this paper, we introduce a Federated Domain-specific Knowledge Transfer (FDKT) framework, which enables domain-specific knowledge transfer from LLMs to SLMs while preserving clients' data privacy. The core insight is to leverage LLMs to augment data based on domain-specific few-shot demonstrations, which are synthesized from private domain data using differential privacy. Such synthetic samples share similar data distribution with clients' private data and allow the server LLM to generate particular knowledge to improve clients' SLMs. The extensive experimental results demonstrate that the proposed FDKT framework consistently and greatly improves SLMs' task performance by around 5\% with a privacy budget of less than 10, compared to local training on private data.

Cross-modal knowledge transfer enhances point cloud representation learning in LiDAR semantic segmentation. Despite its potential, the \textit{weak teacher challenge} arises due to repetitive and non-diverse car camera images and sparse, inaccurate ground truth labels. To address this, we propose the Efficient Image-to-LiDAR Knowledge Transfer (ELiTe) paradigm. ELiTe introduces Patch-to-Point Multi-Stage Knowledge Distillation, transferring comprehensive knowledge from the Vision Foundation Model (VFM), extensively trained on diverse open-world images. This enables effective knowledge transfer to a lightweight student model across modalities. ELiTe employs Parameter-Efficient Fine-Tuning to strengthen the VFM teacher and expedite large-scale model training with minimal costs. Additionally, we introduce the Segment Anything Model based Pseudo-Label Generation approach to enhance low-quality image labels, facilitating robust semantic representations. Efficient knowledge transfer in ELiTe yields state-of-the-art results on the SemanticKITTI benchmark, outperforming real-time inference models. Our approach achieves this with significantly fewer parameters, confirming its effectiveness and efficiency., Comment: 9 pages, 6 figures, ICME 2024 oral

Federated Class-Incremental Learning (FCIL) focuses on continually transferring the previous knowledge to learn new classes in dynamic Federated Learning (FL). However, existing methods do not consider the trustworthiness of FCIL, i.e., improving continual utility, privacy, and efficiency simultaneously, which is greatly influenced by catastrophic forgetting and data heterogeneity among clients. To address this issue, we propose FedProK (Federated Prototypical Feature Knowledge Transfer), leveraging prototypical feature as a novel representation of knowledge to perform spatial-temporal knowledge transfer. Specifically, FedProK consists of two components: (1) feature translation procedure on the client side by temporal knowledge transfer from the learned classes and (2) prototypical knowledge fusion on the server side by spatial knowledge transfer among clients. Extensive experiments conducted in both synchronous and asynchronous settings demonstrate that our FedProK outperforms the other state-of-the-art methods in three perspectives of trustworthiness, validating its effectiveness in selectively transferring spatial-temporal knowledge.

Reinforcement Learning (RL) provides a framework in which agents can be trained, via trial and error, to solve complex decision-making problems. Learning with little supervision causes RL methods to require large amounts of data, which renders them too expensive for many applications (e.g. robotics). By reusing knowledge from a different task, knowledge transfer methods present an alternative to reduce the training time in RL. Given how severe data scarcity can be, there has been a growing interest for methods capable of transferring knowledge across different domains (i.e. problems with different representation) due to the flexibility they offer. This review presents a unifying analysis of methods focused on transferring knowledge across different domains. Through a taxonomy based on a transfer-approach categorization, and a characterization of works based on their data-assumption requirements, the objectives of this article are to 1) provide a comprehensive and systematic revision of knowledge transfer methods for the cross-domain RL setting, 2) categorize and characterize these methods to provide an analysis based on relevant features such as their transfer approach and data requirements, and 3) discuss the main challenges regarding cross-domain knowledge transfer, as well as ideas of future directions worth exploring to address these problems.

In data-driven decision-making in marketing, healthcare, and education, it is desirable to utilize a large amount of data from existing ventures to navigate high-dimensional feature spaces and address data scarcity in new ventures. We explore knowledge transfer in dynamic decision-making by concentrating on batch stationary environments and formally defining task discrepancies through the lens of Markov decision processes (MDPs). We propose a framework of Transferred Fitted $Q$-Iteration algorithm with general function approximation, enabling the direct estimation of the optimal action-state function $Q^*$ using both target and source data. We establish the relationship between statistical performance and MDP task discrepancy under sieve approximation, shedding light on the impact of source and target sample sizes and task discrepancy on the effectiveness of knowledge transfer. We show that the final learning error of the $Q^*$ function is significantly improved from the single task rate both theoretically and empirically.

Large language models (LLMs) and vision language models (VLMs) are changing the world and gradually presenting human-level intelligence in various real-world scenarios, including knowledge-based question answering, mathematics, and programming. During the master period, my research focuses on understanding and improving current large language models’ reasoning capacity towards general problems solving, and efficient methods to enable the knowledge transfer for vision-language models: distill knowledge from large vision-language models.

As Moore's law slows down, computing systems must prioritize higher energy efficiency to sustain performance scaling. GPUs have emerged as the primary workhorses of computing resources, making the achievement of high energy efficiency in GPUs a critical concern. However, implementing runtime power management on GPUs poses significant challenges due to the high variations and complexities arising from workloads and hardware configurations, which render offline optimization and reactive-based methods less effective. In this paper, we present a program counter (PC)-oriented prediction-based power management approach for GPGPUs. Our approach leverages the benefits of prediction to address online variations while enhancing prediction capability through knowledge transfer across different levels of resources. Experiments conducted on realistic applications demonstrate that our proposed method achieves the maximum energy savings under a user-defined performance constraint compared to state-of-the-art designs.

Employing the Foucauldian term ‘conduct’, this article explores how social resilience and morale became a target of state intervention in Denmark during the Cold War. ‘Psychological defence’ was a Cold War phenomenon designed to bring an imagined future war into a space of control as well as a tool for the authorities’ exercise of power in case another world war became a reality. Advocating a methodological internationalism, the article analyses how the concept of psychological defence travelled from Sweden to Denmark via Norway and NATO, and in a complex process of translation, mixing and hybridization was adapted and appropriated to Danish security policy conditions, preparedness culture, and historical experiences. Ultimately, psychological defence was replaced with a more practical or even cynical approach to public information and media preparedness, even if the objectives remained the same. The article employs source material from Danish, Swedish, and NATO archives and combines Scandinavian Cold War history with media history and the history of knowledge.

In this paper, we propose a method to improve the accuracy of speech emotion recognition (SER) by using vision transformer (ViT) to attend to the correlation of frequency (y-axis) with time (x-axis) in spectrogram and transferring positional information between ViT through knowledge transfer. The proposed method has the following originality i) We use vertically segmented patches of log-Mel spectrogram to analyze the correlation of frequencies over time. This type of patch allows us to correlate the most relevant frequencies for a particular emotion with the time they were uttered. ii) We propose the use of image coordinate encoding, an absolute positional encoding suitable for ViT. By normalizing the x, y coordinates of the image to -1 to 1 and concatenating them to the image, we can effectively provide valid absolute positional information for ViT. iii) Through feature map matching, the locality and location information of the teacher network is effectively transmitted to the student network. Teacher network is a ViT that contains locality of convolutional stem and absolute position information through image coordinate encoding, and student network is a structure that lacks positional encoding in the basic ViT structure. In feature map matching stage, we train through the mean absolute error (L1 loss) to minimize the difference between the feature maps of the two networks. To validate the proposed method, three emotion datasets (SAVEE, EmoDB, and CREMA-D) consisting of speech were converted into log-Mel spectrograms for comparison experiments. The experimental results show that the proposed method significantly outperforms the state-of-the-art methods in terms of weighted accuracy while requiring significantly fewer floating point operations (FLOPs). Overall, the proposed method offers an promising solution for SER by providing improved efficiency and performance.

This study explores a learning-based tri-finger robotic arm manipulating task, which requires complex movements and coordination among the fingers. By employing reinforcement learning, we train an agent to acquire the necessary skills for proficient manipulation. To enhance the efficiency and effectiveness of the learning process, two knowledge transfer strategies, fine-tuning and curriculum learning, were utilized within the soft actor-critic architecture. Fine-tuning allows the agent to leverage pre-trained knowledge and adapt it to new tasks. Several variations like model transfer, policy transfer, and across-task transfer were implemented and evaluated. To eliminate the need for pretraining, curriculum learning decomposes the advanced task into simpler, progressive stages, mirroring how humans learn. The number of learning stages, the context of the sub-tasks, and the transition timing were found to be the critical design parameters. The key factors of two learning strategies and corresponding effects were explored in context-aware and context-unaware scenarios, enabling us to identify the scenarios where the methods demonstrate optimal performance, derive conclusive insights, and contribute to a broader range of learning-based engineering applications.

A typical assumption in state-of-the-art self-localization models is that an annotated training dataset is available in the target workspace. However, this does not always hold when a robot travels in a general open-world. This study introduces a novel training scheme for open-world distributed robot systems. In our scheme, a robot ("student") can ask the other robots it meets at unfamiliar places ("teachers") for guidance. Specifically, a pseudo-training dataset is reconstructed from the teacher model and thereafter used for continual learning of the student model. Unlike typical knowledge transfer schemes, our scheme introduces only minimal assumptions on the teacher model, such that it can handle various types of open-set teachers, including uncooperative, untrainable (e.g., image retrieval engines), and blackbox teachers (i.e., data privacy). Rather than relying on the availability of private data of teachers as in existing methods, we propose to exploit an assumption that holds universally in self-localization tasks: "The teacher model is a self-localization system" and to reuse the self-localization system of a teacher as a sole accessible communication channel. We particularly focus on designing an excellent student/questioner whose interactions with teachers can yield effective question-and-answer sequences that can be used as pseudo-training datasets for the student self-localization model. When applied to a generic recursive knowledge distillation scenario, our approach exhibited stable and consistent performance improvement., Comment: 7 pages, 3 figures, technical report

This paper presents a novel framework for continual feature selection (CFS) in data preprocessing, particularly in the context of an open and dynamic environment where unknown classes may emerge. CFS encounters two primary challenges: the discovery of unknown knowledge and the transfer of known knowledge. To this end, the proposed CFS method combines the strengths of continual learning (CL) with granular-ball computing (GBC), which focuses on constructing a granular-ball knowledge base to detect unknown classes and facilitate the transfer of previously learned knowledge for further feature selection. CFS consists of two stages: initial learning and open learning. The former aims to establish an initial knowledge base through multi-granularity representation using granular-balls. The latter utilizes prior granular-ball knowledge to identify unknowns, updates the knowledge base for granular-ball knowledge transfer, reinforces old knowledge, and integrates new knowledge. Subsequently, we devise an optimal feature subset mechanism that incorporates minimal new features into the existing optimal subset, often yielding superior results during each period. Extensive experimental results on public benchmark datasets demonstrate our method's superiority in terms of both effectiveness and efficiency compared to state-of-the-art feature selection methods., Comment: 14 pages, 7 figures, 6 tables

Large-scale vision-language models (VLMs) have shown a strong zero-shot generalization capability on unseen-domain data. However, when adapting pre-trained VLMs to a sequence of downstream tasks, they are prone to forgetting previously learned knowledge and degrade their zero-shot classification capability. To tackle this problem, we propose a unique Selective Dual-Teacher Knowledge Transfer framework that leverages the most recent fine-tuned and the original pre-trained VLMs as dual teachers to preserve the previously learned knowledge and zero-shot capabilities, respectively. With only access to an unlabeled reference dataset, our proposed framework performs a selective knowledge distillation mechanism by measuring the feature discrepancy from the dual teacher VLMs. Consequently, our selective dual-teacher knowledge distillation would mitigate catastrophic forgetting of previously learned knowledge while preserving the zero-shot capabilities from pre-trained VLMs. Through extensive experiments on benchmark datasets, we show that our proposed framework is favorable against state-of-the-art continual learning approaches for preventing catastrophic forgetting and zero-shot degradation.

This paper addresses the challenge of overfitting in the learning of dynamical systems by introducing a novel approach for the generation of synthetic data, aimed at enhancing model generalization and robustness in scenarios characterized by data scarcity. Central to the proposed methodology is the concept of knowledge transfer from systems within the same class. Specifically, synthetic data is generated through a pre-trained meta-model that describes a broad class of systems to which the system of interest is assumed to belong. Training data serves a dual purpose: firstly, as input to the pre-trained meta model to discern the system's dynamics, enabling the prediction of its behavior and thereby generating synthetic output sequences for new input sequences; secondly, in conjunction with synthetic data, to define the loss function used for model estimation. A validation dataset is used to tune a scalar hyper-parameter balancing the relative importance of training and synthetic data in the definition of the loss function. The same validation set can be also used for other purposes, such as early stopping during the training, fundamental to avoid overfitting in case of small-size training datasets. The efficacy of the approach is shown through a numerical example that highlights the advantages of integrating synthetic data into the system identification process.