Search

Deep learning models, in particular \textit{image} models, have recently gained generalisability and robustness. %are becoming more general and robust by the day. In this work, we propose to exploit such advances in the realm of \textit{video} classification. Video foundation models suffer from the requirement of extensive pretraining and a large training time. Towards mitigating such limitations, we propose "\textit{Attention Map (AM) Flow}" for image models, a method for identifying pixels relevant to motion in each input video frame. In this context, we propose two methods to compute AM flow, depending on camera motion. AM flow allows the separation of spatial and temporal processing, while providing improved results over combined spatio-temporal processing (as in video models). Adapters, one of the popular techniques in parameter efficient transfer learning, facilitate the incorporation of AM flow into pretrained image models, mitigating the need for full-finetuning. We extend adapters to "\textit{temporal processing adapters}" by incorporating a temporal processing unit into the adapters. Our work achieves faster convergence, therefore reducing the number of epochs needed for training. Moreover, we endow an image model with the ability to achieve state-of-the-art results on popular action recognition datasets. This reduces training time and simplifies pretraining. We present experiments on Kinetics-400, Something-Something v2, and Toyota Smarthome datasets, showcasing state-of-the-art or comparable results.

Deep Neural Networks (DNNs) suffer from significant retraining costs when adapting to evolving requirements. Modularizing DNNs offers the promise of improving their reusability. Previous work has proposed techniques to decompose DNN models into modules both during and after training. However, these strategies yield several shortcomings, including significant weight overlaps and accuracy losses across modules, restricted focus on convolutional layers only, and added complexity and training time by introducing auxiliary masks to control modularity. In this work, we propose MODA, an activation-driven modular training approach. MODA promotes inherent modularity within a DNN model by directly regulating the activation outputs of its layers based on three modular objectives: intra-class affinity, inter-class dispersion, and compactness. MODA is evaluated using three well-known DNN models and three datasets with varying sizes. This evaluation indicates that, compared to the existing state-of-the-art, using MODA yields several advantages: (1) MODA accomplishes modularization with 29% less training time; (2) the resultant modules generated by MODA comprise 2.4x fewer weights and 3.5x less weight overlap while (3) preserving the original model's accuracy without additional fine-tuning; in module replacement scenarios, (4) MODA improves the accuracy of a target class by 12% on average while ensuring minimal impact on the accuracy of other classes.

Code translation transforms programs from one programming language (PL) to another. Several rule-based transpilers have been designed to automate code translation between different pairs of PLs. However, the rules can become obsolete as the PLs evolve and cannot generalize to other PLs. Recent studies have explored the automation of code translation using Large Language Models (LLMs). One key observation is that such techniques may work well for crafted benchmarks but fail to generalize to the scale and complexity of real-world projects with dependencies, custom types, PL-specific features, etc. We propose AlphaTrans, a neuro-symbolic approach to automate repository-level code translation. AlphaTrans translates both source and test code, and employs multiple levels of validation to ensure the translation preserves the functionality of the source program. To break down the problem for LLMs, AlphaTrans leverages program analysis to decompose the program into fragments and translates them in the reverse call order. We leveraged AlphaTrans to translate ten real-world open-source projects consisting of <836, 8575, 2719> classes, methods, and tests. AlphaTrans translated the entire repository of these projects consisting of 6899 source code fragments. 99.1% of the translated code fragments are syntactically correct, and AlphaTrans validates the translations' runtime behavior and functional correctness for 25.8%. On average, the integrated translation and validation take 36 hours to translate a project, showing its scalability in practice. For the syntactically or semantically incorrect translations, AlphaTrans generates a report including existing translation, stack trace, test errors, or assertion failures. We provided these artifacts to two developers to fix the translation bugs in four projects. They were able to fix the issues in 20.1 hours on average and achieve all passing tests.

This paper presents the development of machine learning (ML) models to predict hypoxemia severity during emergency triage, especially in Chemical, Biological, Radiological, Nuclear, and Explosive (CBRNE) events, using physiological data from medical-grade sensors. Gradient Boosting Models (XGBoost, LightGBM, CatBoost) and sequential models (LSTM, GRU) were trained on physiological and demographic data from the MIMIC-III and IV datasets. A robust preprocessing pipeline addressed missing data, class imbalances, and incorporated synthetic data flagged with masks. Gradient Boosting Models (GBMs) outperformed sequential models in terms of training speed, interpretability, and reliability, making them well-suited for real-time decision-making. While their performance was comparable to that of sequential models, the GBMs used score features from six physiological variables derived from the enhanced National Early Warning Score (NEWS) 2, which we termed NEWS2+. This approach significantly improved prediction accuracy. While sequential models handled temporal data well, their performance gains did not justify the higher computational cost. A 5-minute prediction window was chosen for timely intervention, with minute-level interpolations standardizing the data. Feature importance analysis highlighted the significant role of mask and score features in enhancing both transparency and performance. Temporal dependencies proved to be less critical, as Gradient Boosting Models were able to capture key patterns effectively without relying on them. This study highlights ML's potential to improve triage and reduce alarm fatigue. Future work will integrate data from multiple hospitals to enhance model generalizability across clinical settings., Comment: 12 figures, 12 tables and 39 pages

We present a stochastic model for two successive SIR (Susceptible, Infectious, Recovered) epidemics in the same network structured population. Individuals infected during the first epidemic might have (partial) immunity for the second one. The first epidemic is analysed through a bond percolation model, while the second epidemic is approximated by a three-type branching process in which the types of individuals depend on their position in the percolation clusters used for the first epidemic. This branching process approximation enables us to calculate, in the large population limit and conditional upon a large outbreak in the first epidemic, a threshold parameter and the probability of a large outbreak for the second epidemic. A second branching process approximation enables us to calculate the fraction of the population that are infected by such a second large outbreak. We illustrate our results through some specific cases which have appeared previously in the literature and show that our asymptotic results give good approximations for finite populations., Comment: 45 pages, 5 figures

The computer vision community has explored dyadic interactions for atomic actions such as pushing, carrying-object, etc. However, with the advancement in deep learning models, there is a need to explore more complex dyadic situations such as loose interactions. These are interactions where two people perform certain atomic activities to complete a global action irrespective of temporal synchronisation and physical engagement, like cooking-together for example. Analysing these types of dyadic-interactions has several useful applications in the medical domain for social-skills development and mental health diagnosis. To achieve this, we propose a novel dual-path architecture to capture the loose interaction between two individuals. Our model learns global abstract features from each stream via a CNNs backbone and fuses them using a new Global-Layer-Attention module based on a cross-attention strategy. We evaluate our model on real-world autism diagnoses such as our Loose-Interaction dataset, and the publicly available Autism dataset for loose interactions. Our network achieves baseline results on the Loose-Interaction and SOTA results on the Autism datasets. Moreover, we study different social interactions by experimenting on a publicly available dataset i.e. NTU-RGB+D (interactive classes from both NTU-60 and NTU-120). We have found that different interactions require different network designs. We also compare a slightly different version of our method by incorporating time information to address tight interactions achieving SOTA results.

Crack detection, particularly from pavement images, presents a formidable challenge in the domain of computer vision due to several inherent complexities such as intensity inhomogeneity, intricate topologies, low contrast, and noisy backgrounds. Automated crack detection is crucial for maintaining the structural integrity of essential infrastructures, including buildings, pavements, and bridges. Existing lightweight methods often face challenges including computational inefficiency, complex crack patterns, and difficult backgrounds, leading to inaccurate detection and impracticality for real-world applications. To address these limitations, we propose EfficientCrackNet, a lightweight hybrid model combining Convolutional Neural Networks (CNNs) and transformers for precise crack segmentation. EfficientCrackNet integrates depthwise separable convolutions (DSC) layers and MobileViT block to capture both global and local features. The model employs an Edge Extraction Method (EEM) and for efficient crack edge detection without pretraining, and Ultra-Lightweight Subspace Attention Module (ULSAM) to enhance feature extraction. Extensive experiments on three benchmark datasets Crack500, DeepCrack, and GAPs384 demonstrate that EfficientCrackNet achieves superior performance compared to existing lightweight models, while requiring only 0.26M parameters, and 0.483 FLOPs (G). The proposed model offers an optimal balance between accuracy and computational efficiency, outperforming state-of-the-art lightweight models, and providing a robust and adaptable solution for real-world crack segmentation.

The accurate first-principles description of strongly-correlated materials is an important and challenging problem in condensed matter physics. Ab initio downfolding has emerged as a way of deriving accurate many-body Hamiltonians including strong correlations, representing a subspace of interest of a material, using density functional theory calculations as a starting point. However, the solution of these material-specific models can scale exponentially on classical computers, constituting a challenge. Here we propose that utilizing quantum computers for obtaining the properties of downfolded Hamiltonians yields an accurate description of the ground state properties of strongly-correlated systems, while circumventing the exponential scaling problem. We demonstrate this for diverse strongly-correlated materials by combining ab initio downfolding and variational quantum eigensolvers, correctly predicting the antiferromagnetic state of one-dimensional cuprate $\text{Ca}_2\text{CuO}_3$, the excitonic ground state of monolayer $\text{WTe}_2$, and the charge-ordered state of correlated metal $\text{SrVO}_3$. By utilizing a classical tensor network implementation of variational quantum eigensolvers we are able to simulate large models with up to $54$ qubits and encompassing up to four bands in the correlated subspace, which is indicative of the complexity that our framework can address.

Accurate prediction of house price, a vital aspect of the residential real estate sector, is of substantial interest for a wide range of stakeholders. However, predicting house prices is a complex task due to the significant variability influenced by factors such as house features, location, neighborhood, and many others. Despite numerous attempts utilizing a wide array of algorithms, including recent deep learning techniques, to predict house prices accurately, existing approaches have fallen short of considering a wide range of factors such as textual and visual features. This paper addresses this gap by comprehensively incorporating attributes, such as features, textual descriptions, geo-spatial neighborhood, and house images, typically showcased in real estate listings in a house price prediction system. Specifically, we propose a multi-modal deep learning approach that leverages different types of data to learn more accurate representation of the house. In particular, we learn a joint embedding of raw house attributes, geo-spatial neighborhood, and most importantly from textual description and images representing the house; and finally use a downstream regression model to predict the house price from this jointly learned embedding vector. Our experimental results with a real-world dataset show that the text embedding of the house advertisement description and image embedding of the house pictures in addition to raw attributes and geo-spatial embedding, can significantly improve the house price prediction accuracy. The relevant source code and dataset are publicly accessible at the following URL: https://github.com/4P0N/mhpp, Comment: 22 pages

Let $G(\mathbb{Q})$ be a simply connected Chevalley group over $\mathbb{Q}$ corresponding to a simple Lie algebra $\mathfrak g$ over $\mathbb{C}$. Let $V$ be a finite dimensional faithful highest weight $\mathfrak g$-module and let $V_\mathbb{Z}$ be a Chevalley $\mathbb{Z}$-form of $V$. Let $\Gamma(\mathbb{Z})$ be the subgroup of $G(\mathbb{Q})$ that preserves $V_{\mathbb{Z}}$ and let $G(\mathbb{Z})$ be the group of $\mathbb{Z}$-points of $G(\mathbb{Q})$. Then $G(\mathbb{Q})$ is \emph{integral} if $G(\mathbb{Z})=\Gamma(\mathbb{Z})$. Chevalley's original work constructs a scheme-theoretic integral form of $G(\mathbb{Q})$ which equals $\Gamma(\mathbb{Z})$. Here we give a representation-theoretic proof of integrality of $G(\mathbb{Q})$ using only the action of $G(\mathbb{Q})$ on $V$, rather than the language of group schemes. We discuss the challenges and open problems that arise in trying to extend this to a proof of integrality for Kac-Moody groups over $\mathbb{Q}$.

Global environmental challenges and rising energy demands have led to extensive exploration of wind energy technologies. Accurate wind speed forecasting (WSF) is crucial for optimizing wind energy capture and ensuring system stability. However, predicting wind speed remains challenging due to its inherent randomness, fluctuation, and unpredictability. This study proposes the Temporal Convolutional Network Former (TCNFormer) for short-term (12-hour) wind speed forecasting. The TCNFormer integrates the Temporal Convolutional Network (TCN) and transformer encoder to capture the spatio-temporal features of wind speed. The transformer encoder consists of two distinct attention mechanisms: causal temporal multi-head self-attention (CT-MSA) and temporal external attention (TEA). CT-MSA ensures that the output of a step derives only from previous steps, i.e., causality. Locality is also introduced to improve efficiency. TEA explores potential relationships between different sample sequences in wind speed data. This study utilizes wind speed data from the NASA Prediction of Worldwide Energy Resources (NASA POWER) of Patenga Sea Beach, Chittagong, Bangladesh (latitude 22.2352{\deg} N, longitude 91.7914{\deg} E) over a year (six seasons). The findings indicate that the TCNFormer outperforms state-of-the-art models in prediction accuracy. The proposed TCNFormer presents a promising method for spatio-temporal WSF and may achieve desirable performance in real-world applications of wind power systems.

Bio-engineered robots are under rapid development due to their maneuver ability through uneven surfaces. This advancement paves the way for experimenting with versatile electrical system developments with various motors. In this research paper, we present a design, fabrication and analysis of a versatile printed circuit board (PCB) as the main system that allows for the control of twelve stepper motors by stacking low-budget stepper motor controller and widely used micro-controller unit. The primary motivation behind the design is to offer a compact and efficient hardware solution for controlling multiple stepper motors of a quadruped robot while meeting the required power budget. The research focuses on the hardware's architecture, stackable design, power budget planning and a thorough analysis. Additionally, PDN (Power Distribution Network) analysis simulation is done to ensure that the voltage and current density are within the expected parameters. Also, the hardware design deep dives into design for manufacturability (DFM). The ability to stack the controllers on the development board provides insights into the board's components swapping feasibility. The findings from this research make a significant contribution to the advancement of stepper motor control systems of multi-axis applications for bio-inspired robot offering a convenient form factor and a reliable performance., Comment: LaTeX, 26 pages with 25 figures

This paper presents the External Attention Vision Transformer (EAViT) model, a novel approach designed to enhance audio classification accuracy. As digital audio resources proliferate, the demand for precise and efficient audio classification systems has intensified, driven by the need for improved recommendation systems and user personalization in various applications, including music streaming platforms and environmental sound recognition. Accurate audio classification is crucial for organizing vast audio libraries into coherent categories, enabling users to find and interact with their preferred audio content more effectively. In this study, we utilize the GTZAN dataset, which comprises 1,000 music excerpts spanning ten diverse genres. Each 30-second audio clip is segmented into 3-second excerpts to enhance dataset robustness and mitigate overfitting risks, allowing for more granular feature analysis. The EAViT model integrates multi-head external attention (MEA) mechanisms into the Vision Transformer (ViT) framework, effectively capturing long-range dependencies and potential correlations between samples. This external attention (EA) mechanism employs learnable memory units that enhance the network's capacity to process complex audio features efficiently. The study demonstrates that EAViT achieves a remarkable overall accuracy of 93.99%, surpassing state-of-the-art models.

The circular economy has the potential to increase resource efficiency and minimize waste through the 4R framework of reducing, reusing, recycling, and recovering. Blockchain technology is currently considered a valuable aid in the transition to a circular economy. Its decentralized and tamper-resistant nature enables the construction of transparent and secure supply chain management systems, thereby improving product accountability and traceability. However, the full potential of blockchain technology in circular economy models will not be realized until a number of concerns, including scalability, interoperability, data protection, and regulatory and legal issues, are addressed. More research and stakeholder participation are required to overcome these limitations and achieve the benefits of blockchain technology in promoting a circular economy. This article presents a systematic literature review (SLR) that identified industry use cases for blockchain-driven circular economy models and offered architectures to minimize resource consumption, prices, and inefficiencies while encouraging the reuse, recycling, and recovery of end-of-life products. Three main outcomes emerged from our review of 41 documents, which included scholarly publications, Twitter-linked information, and Google results. The relationship between blockchain and the 4R framework for circular economy; discussion the terminology and various forms of blockchain and circular economy; and identification of the challenges and obstacles that blockchain technology may face in enabling a circular economy. This research shows how blockchain technology can help with the transition to a circular economy. Yet, it emphasizes the importance of additional study and stakeholder participation to overcome potential hurdles and obstacles in implementing blockchain-driven circular economy models.

Disaster events often unfold rapidly, necessitating a swift and effective response. Developing action plans, resource allocation, and resolution of help requests in disaster scenarios is time-consuming and complex since disaster-relevant information is often uncertain. Leveraging real-time data can significantly deal with data uncertainty and enhance disaster response efforts. To deal with real-time data uncertainty, social media appeared as an alternative effective source of real-time data as there has been extensive use of social media during and after the disasters. However, it also brings forth challenges regarding trustworthiness and bias in these data. To fully leverage social media data for disaster management, it becomes crucial to mitigate biases that may arise due to specific disaster types or regional contexts. Additionally, the presence of misinformation within social media data raises concerns about the reliability of data sources, potentially impeding actionable insights and leading to improper resource utilization. To overcome these challenges, our research aimed to investigate how to ensure trustworthiness and address biases in social media data. We aim to investigate and identify the factors that can be used to enhance trustworthiness and minimize bias to make an efficient and scalable disaster management system utilizing real-time social media posts, identify disaster-related keywords, and assess the severity of the disaster. By doing so, the integration of real-time social data can improve the speed and accuracy of disaster management systems, Comment: Submitted abstract at Yorkshire Innovation in Science and Engineering (YISEC) 2024

Simulating the Hubbard model is of great interest to a wide range of applications within condensed matter physics, however its solution on classical computers remains challenging in dimensions larger than one. The relative simplicity of this model, embodied by the sparseness of the Hamiltonian matrix, allows for its efficient implementation on quantum computers, and for its approximate solution using variational algorithms such as the variational quantum eigensolver. While these algorithms have been shown to reproduce the qualitative features of the Hubbard model, their quantitative accuracy in terms of producing true ground state energies and other properties, and the dependence of this accuracy on the system size and interaction strength, the choice of variational ansatz, and the degree of spatial inhomogeneity in the model, remains unknown. Here we present a rigorous classical benchmarking study, demonstrating the potential impact of these factors on the accuracy of the variational solution of the Hubbard model on quantum hardware. We find that even when using the most accurate wavefunction ans\"{a}tze for the Hubbard model, the error in its ground state energy and wavefunction plateaus for larger lattices, while stronger electronic correlations magnify this issue. Concurrently, spatially inhomogeneous parameters and the presence of off-site Coulomb interactions only have a small effect on the accuracy of the computed ground state energies. Our study highlights the capabilities and limitations of current approaches for solving the Hubbard model on quantum hardware, and we discuss potential future avenues of research.

Machine learning (ML) models have emerged as powerful tools for accelerating materials discovery and design by enabling accurate predictions of properties from compositional and structural data. These capabilities are vital for developing advanced technologies across fields such as energy, electronics, and biomedicine, potentially reducing the time and resources needed for new material exploration and promoting rapid innovation cycles. Recent efforts have focused on employing advanced ML algorithms, including deep learning - based graph neural network, for property prediction. Additionally, ensemble models have proven to enhance the generalizability and robustness of ML and DL. However, the use of such ensemble strategies in deep graph networks for material property prediction remains underexplored. Our research provides an in-depth evaluation of ensemble strategies in deep learning - based graph neural network, specifically targeting material property prediction tasks. By testing the Crystal Graph Convolutional Neural Network (CGCNN) and its multitask version, MT-CGCNN, we demonstrated that ensemble techniques, especially prediction averaging, substantially improve precision beyond traditional metrics for key properties like formation energy per atom ($\Delta E^{f}$), band gap ($E_{g}$) and density ($\rho$) in 33,990 stable inorganic materials. These findings support the broader application of ensemble methods to enhance predictive accuracy in the field., Comment: 9 pages, 6 figures, 2 tables

We introduce symplectic left Leibniz algebras and symplectic right Leibniz algebras as generalizations of symplectic Lie algebras. These algebras possess a left symmetric product and are Lie-admissible. We describe completely symmetric Leibniz algebras that are symplectic as both left and right Leibniz algebras. Additionally, we show that symplectic left or right Leibniz algebras can be constructed from a symplectic Lie algebra and a vector space through a method that combines the double extension process and the $T^*$-extension. This approach allows us to generate a broad class of examples., Comment: 28 pages, submitted

Graph transformers are a recent advancement in machine learning, offering a new class of neural network models for graph-structured data. The synergy between transformers and graph learning demonstrates strong performance and versatility across various graph-related tasks. This survey provides an in-depth review of recent progress and challenges in graph transformer research. We begin with foundational concepts of graphs and transformers. We then explore design perspectives of graph transformers, focusing on how they integrate graph inductive biases and graph attention mechanisms into the transformer architecture. Furthermore, we propose a taxonomy classifying graph transformers based on depth, scalability, and pre-training strategies, summarizing key principles for effective development of graph transformer models. Beyond technical analysis, we discuss the applications of graph transformer models for node-level, edge-level, and graph-level tasks, exploring their potential in other application scenarios as well. Finally, we identify remaining challenges in the field, such as scalability and efficiency, generalization and robustness, interpretability and explainability, dynamic and complex graphs, as well as data quality and diversity, charting future directions for graph transformer research., Comment: 23 pages, 4 figures

Autism Spectrum Disorder (ASD) is a diverse collection of neurobiological conditions marked by challenges in social communication and reciprocal interactions, as well as repetitive and stereotypical behaviors. Atypical behavior patterns in a long, untrimmed video can serve as biomarkers for children with ASD. In this paper, we propose a video-based weakly-supervised method that takes spatio-temporal features of long videos to learn typical and atypical behaviors for autism detection. On top of that, we propose a shallow TCN-MLP network, which is designed to further categorize the severity score. We evaluate our method on actual evaluation videos of children with autism collected and annotated (for severity score) by clinical professionals. Experimental results demonstrate the effectiveness of behavioral biomarkers that could help clinicians in autism spectrum analysis.

As opposed to evaluating computation and logic-based reasoning, current benchmarks for evaluating large language models (LLMs) in medicine are primarily focused on question-answering involving domain knowledge and descriptive reasoning. While such qualitative capabilities are vital to medical diagnosis, in real-world scenarios, doctors frequently use clinical calculators that follow quantitative equations and rule-based reasoning paradigms for evidence-based decision support. To this end, we propose MedCalc-Bench, a first-of-its-kind dataset focused on evaluating the medical calculation capability of LLMs. MedCalc-Bench contains an evaluation set of over 1000 manually reviewed instances from 55 different medical calculation tasks. Each instance in MedCalc-Bench consists of a patient note, a question requesting to compute a specific medical value, a ground truth answer, and a step-by-step explanation showing how the answer is obtained. While our evaluation results show the potential of LLMs in this area, none of them are effective enough for clinical settings. Common issues include extracting the incorrect entities, not using the correct equation or rules for a calculation task, or incorrectly performing the arithmetic for the computation. We hope our study highlights the quantitative knowledge and reasoning gaps in LLMs within medical settings, encouraging future improvements of LLMs for various clinical calculation tasks., Comment: Github link: https://github.com/ncbi-nlp/MedCalc-Bench HuggingFace link: https://huggingface.co/datasets/nsk7153/MedCalc-Bench

The ability to perform ab initio molecular dynamics simulations using potential energies calculated on quantum computers would allow virtually exact dynamics for chemical and biochemical systems, with substantial impacts on the fields of catalysis and biophysics. However, noisy hardware, the costs of computing gradients, and the number of qubits required to simulate large systems present major challenges to realizing the potential of dynamical simulations using quantum hardware. Here, we demonstrate that some of these issues can be mitigated by recent advances in machine learning. By combining transfer learning with techniques for building machine-learned potential energy surfaces, we propose a new path forward for molecular dynamics simulations on quantum hardware. We use transfer learning to reduce the number of energy evaluations that use quantum hardware by first training models on larger, less accurate classical datasets and then refining them on smaller, more accurate quantum datasets. We demonstrate this approach by training machine learning models to predict a molecule's potential energy using Behler-Parrinello neural networks. When successfully trained, the model enables energy gradient predictions necessary for dynamics simulations that cannot be readily obtained directly from quantum hardware. To reduce the quantum resources needed, the model is initially trained with data derived from low-cost techniques, such as Density Functional Theory, and subsequently refined with a smaller dataset obtained from the optimization of the Unitary Coupled Cluster ansatz. We show that this approach significantly reduces the size of the quantum training dataset while capturing the high accuracies needed for quantum chemistry simulations., Comment: 1- pages, 12 figures

Facial Recognition is a technique, based on machine learning technology that can recognize a human being analyzing his facial profile, and is applied in solving various types of realworld problems nowadays. In this paper, a common real-world problem, finding a missing person has been solved in a secure and effective way with the help of facial recognition technology. Although there exist a few works on solving the problem, the proposed work is unique with respect to its security, design, and feasibility. Impeding intruders in participating in the processes and giving importance to both finders and family members of a missing person are two of the major features of this work. The proofs of the works of our system in finding a missing person have been described in the result section of the paper. The advantages that our system provides over the other existing systems can be realized from the comparisons, described in the result summary section of the paper. The work is capable of providing a worthy solution to find a missing person on the digital platform.

Leveraging large-scale pre-training, vision foundational models showcase notable performance benefits. Recent segmentation algorithms for natural scenes have advanced significantly. However, existing models still struggle to automatically segment personalized instances in dense and crowded scenarios, where severe occlusions, scale variations, and background clutter pose a challenge to accurately delineate densely packed instances of the target object. To address this, we propose PerSense, an end-to-end, training-free, and model-agnostic one-shot framework for Personalized instance Segmentation in dense images. We develop a new baseline capable of automatically generating instance-level point prompts via proposing a novel Instance Detection Module (IDM) that leverages density maps, encapsulating spatial distribution of objects in an image. To mitigate false positives within generated point prompts, we design Point Prompt Selection Module (PPSM). Both IDM and PPSM transform density maps into personalized precise point prompts for instance-level segmentation and offer a seamless integration in our model-agnostic framework. We also introduce a feedback mechanism which enables PerSense to improve the accuracy of density maps by automating the exemplar selection process for density map generation. To promote algorithmic advances and effective tools for this relatively underexplored task, we introduce PerSense-D, a diverse dataset exclusive to personalized instance segmentation in dense images. Our extensive experiments establish PerSense superiority in dense scenarios by achieving an mIoU of 71.61% on PerSense-D, outperforming recent SOTA models by significant margins of +47.16%, +42.27%, +8.83%, and +5.69%. Additionally, our qualitative findings demonstrate the adaptability of our framework to images captured in-the-wild., Comment: Technical report of PerSense

The spread of microbial infections is governed by the self-organization of bacteria on surfaces. Limitations of live imaging techniques make collective behaviors in clinically relevant systems challenging to quantify. Here, novel experimental and image analysis techniques for high-fidelity single-cell segmentation of bacterial colonies are developed. Machine learning-based segmentation models are trained solely using synthetic microscopy images that are processed to look realistic using state-of-the-art image-to-image translation methods, requiring no biophysical modeling. Accurate single-cell segmentation is achieved for densely packed single-species colonies and multi-species colonies of common pathogenic bacteria, even under suboptimal imaging conditions and for both brightfield and confocal laser scanning microscopy. The resulting data provide quantitative insights into the self-organization of bacteria on soft surfaces. Thanks to their high adaptability and relatively simple implementation, these methods promise to greatly facilitate quantitative descriptions of bacterial infections in varied environments., Comment: 12 pages, 5 figures

GenAI has gained the attention of a myriad of users in almost every profession. Its advancement has had an intense impact on education, significantly disrupting the assessment design and evaluation methodologies. Despite the potential benefits and possibilities of GenAI in the education sector, there are several concerns primarily centred around academic integrity, authenticity, equity of access, assessment evaluation methodology, and feedback. Consequently, academia is encountering challenges in assessment design that are essential to retaining academic integrity in the age of GenAI. In this article, we discuss the challenges, and opportunities that need to be addressed for the assessment design and evaluation. The article also highlights the importance of clear policy about the usage of GenAI in completing assessment tasks, and also in design approaches to ensure academic integrity and subject learning. Additionally, this article also provides assessment categorisation based on the use of GenAI to cultivate knowledge among students and academic professionals. It also provides information on the skills necessary to formulate and articulate problems and evaluate the task, enabling students and academics to effectively utilise GenAI tools., Comment: 5 pages, 2 figures

Low Rank Adaptation (LoRA) has emerged as one of the most widely adopted methods for Parameter Efficient Fine-Tuning (PEFT) of Large Language Models (LLMs). LoRA reduces the number of trainable parameters and memory usage while achieving comparable performance to full fine-tuning. We aim to assess the viability of training and serving LLMs fine-tuned with LoRA in real-world applications. First, we measure the quality of LLMs fine-tuned with quantized low rank adapters across 10 base models and 31 tasks for a total of 310 models. We find that 4-bit LoRA fine-tuned models outperform base models by 34 points and GPT-4 by 10 points on average. Second, we investigate the most effective base models for fine-tuning and assess the correlative and predictive capacities of task complexity heuristics in forecasting the outcomes of fine-tuning. Finally, we evaluate the latency and concurrency capabilities of LoRAX, an open-source Multi-LoRA inference server that facilitates the deployment of multiple LoRA fine-tuned models on a single GPU using shared base model weights and dynamic adapter loading. LoRAX powers LoRA Land, a web application that hosts 25 LoRA fine-tuned Mistral-7B LLMs on a single NVIDIA A100 GPU with 80GB memory. LoRA Land highlights the quality and cost-effectiveness of employing multiple specialized LLMs over a single, general-purpose LLM.

Complex systems often show macroscopic coherent behavior due to the interactions of microscopic agents like molecules, cells, or individuals in a population with their environment. However, simulating such systems poses several computational challenges during simulation as the underlying dynamics vary and span wide spatiotemporal scales of interest. To capture the fast-evolving features, finer time steps are required while ensuring that the simulation time is long enough to capture the slow-scale behavior, making the analyses computationally unmanageable. This paper showcases how deep learning techniques can be used to develop a precise time-stepping approach for multiscale systems using the joint discovery of coordinates and flow maps. While the former allows us to represent the multiscale dynamics on a representative basis, the latter enables the iterative time-stepping estimation of the reduced variables. The resulting framework achieves state-of-the-art predictive accuracy while incurring lesser computational costs. We demonstrate this ability of the proposed scheme on the large-scale Fitzhugh Nagumo neuron model and the 1D Kuramoto-Sivashinsky equation in the chaotic regime.

Adaptive quantum circuits, which combine local unitary gates, midcircuit measurements, and feedforward operations, have recently emerged as a promising avenue for efficient state preparation, particularly on near-term quantum devices limited to shallow-depth circuits. Matrix product states (MPS) comprise a significant class of many-body entangled states, efficiently describing the ground states of one-dimensional gapped local Hamiltonians and finding applications in a number of recent quantum algorithms. Recently, it was shown that the AKLT state -- a paradigmatic example of an MPS -- can be exactly prepared with an adaptive quantum circuit of constant-depth, an impossible feat with local unitary gates due to its nonzero correlation length [Smith et al., PRX Quantum 4, 020315 (2023)]. In this work, we broaden the scope of this approach and demonstrate that a diverse class of MPS can be exactly prepared using constant-depth adaptive quantum circuits, outperforming optimal preparation protocols that rely on unitary circuits alone. We show that this class includes short- and long-ranged entangled MPS, symmetry-protected topological (SPT) and symmetry-broken states, MPS with finite Abelian, non-Abelian, and continuous symmetries, resource states for MBQC, and families of states with tunable correlation length. Moreover, we illustrate the utility of our framework for designing constant-depth sampling protocols, such as for random MPS or for generating MPS in a particular SPT phase. We present sufficient conditions for particular MPS to be preparable in constant time, with global on-site symmetry playing a pivotal role. Altogether, this work demonstrates the immense promise of adaptive quantum circuits for efficiently preparing many-body entangled states and provides explicit algorithms that outperform known protocols to prepare an essential class of states., Comment: 25 pages, 5 figures