Search

Cooperative grasping and transportation require effective coordination to complete the task. This study focuses on the approach leveraging force-sensing feedback, where robots use sensors to detect forces applied by others on an object to achieve coordination. Unlike explicit communication, it avoids delays and interruptions; however, force-sensing is highly sensitive and prone to interference from variations in grasping environment, such as changes in grasping force, grasping pose, object size and geometry, which can interfere with force signals, subsequently undermining coordination. We propose multi-agent reinforcement learning (MARL) with ternary force representation, a force representation that maintains consistent representation against variations in grasping environment. The simulation and real-world experiments demonstrate the robustness of the proposed method to changes in grasping force, object size and geometry as well as inherent sim2real gap.

Large Language Models (LLMs) show potential for medical applications but often lack specialized clinical knowledge. Retrieval Augmented Generation (RAG) allows customization with domain-specific information, making it suitable for healthcare. This study evaluates the accuracy, consistency, and safety of RAG models in determining fitness for surgery and providing preoperative instructions. We developed LLM-RAG models using 35 local and 23 international preoperative guidelines and tested them against human-generated responses. A total of 3,682 responses were evaluated. Clinical documents were processed using Llamaindex, and 10 LLMs, including GPT3.5, GPT4, and Claude-3, were assessed. Fourteen clinical scenarios were analyzed, focusing on seven aspects of preoperative instructions. Established guidelines and expert judgment were used to determine correct responses, with human-generated answers serving as comparisons. The LLM-RAG models generated responses within 20 seconds, significantly faster than clinicians (10 minutes). The GPT4 LLM-RAG model achieved the highest accuracy (96.4% vs. 86.6%, p=0.016), with no hallucinations and producing correct instructions comparable to clinicians. Results were consistent across both local and international guidelines. This study demonstrates the potential of LLM-RAG models for preoperative healthcare tasks, highlighting their efficiency, scalability, and reliability., Comment: arXiv admin note: substantial text overlap with arXiv:2402.01733

Palmprint recognition has emerged as a prominent biometric authentication method, owing to its high discriminative power and user-friendly nature. This paper introduces a novel Cross-Chirality Palmprint Verification (CCPV) framework that challenges the conventional wisdom in traditional palmprint verification systems. Unlike existing methods that typically require storing both left and right palmprints, our approach enables verification using either palm while storing only one palmprint template. The core of our CCPV framework lies in a carefully designed matching rule. This rule involves flipping both the gallery and query palmprints and calculating the average distance between each pair as the final matching distance. This approach effectively reduces matching variance and enhances overall system robustness. We introduce a novel cross-chirality loss function to construct a discriminative and robust cross-chirality feature space. This loss enforces representation consistency across four palmprint variants: left, right, flipped left, and flipped right. The resulting compact feature space, coupled with the model's enhanced discriminative representation capability, ensures robust performance across various scenarios. We conducted extensive experiments to validate the efficacy of our proposed method. The evaluation encompassed multiple public datasets and considered both closed-set and open-set settings. The results demonstrate the CCPV framework's effectiveness and highlight its potential for real-world applications in palmprint authentication systems.

We introduce an even-parity precession protocol that can detect the nonclassicality of some quantum states using only measurements of a uniformly-precessing variable at different points in time. Depending on the system under study, the protocol may detect the Wigner negativity of a single quantum harmonic oscillator or of a single spin $j\geq 2$; the non-Gaussian entanglement of two harmonic oscillators; or genuine multipartite entanglement of a spin ensemble, whose total spin is an integer. Unlike in other nonclassicality tests, simultaneous or sequential measurements are not required. Our protocol can also detect states that commute with the parity operator, which were missed by similar protocols built from Tsirelson's original precession protocol. This work also closes a long-standing gap by showing the possibility of detecting the Greenberger-Horne-Zeilinger entanglement of an even number of qubits using only collective spin measurements., Comment: 12 pages, 9 figures (final published version)

The COVID-19 pandemic and the implementation of social distancing policies have rapidly changed people's visiting patterns, as reflected in mobility data that tracks mobility traffic using location trackers on cell phones. However, the frequency and duration of concurrent occupancy at specific locations govern the transmission rather than the number of customers visiting. Therefore, understanding how people interact in different locations is crucial to target policies, inform contact tracing, and prevention strategies. This study proposes an efficient way to reduce the spread of the virus among on-campus university students by developing a self-developed Google History Location Extractor and Indicator software based on real-world human mobility data. The platform enables policymakers and researchers to explore the possibility of future developments in the epidemic's spread and simulate the outcomes of human mobility and epidemic state under different epidemic control policies. It offers functions for determining potential contacts, assessing individual infection risks, and evaluating the effectiveness of on-campus policies. The proposed multi-functional platform facilitates the screening process by more accurately targeting potential virus carriers and aids in making informed decisions on epidemic control policies, ultimately contributing to preventing and managing future outbreaks., Comment: 8 pages

Vision-language (VL) models, pretrained on colossal image-text datasets, have attained broad VL competence that is difficult to evaluate. A common belief is that a small number of VL skills underlie the variety of VL tests. In this paper, we perform a large-scale transfer learning experiment aimed at discovering latent VL skills from data. We reveal interesting characteristics that have important implications for test suite design. First, generation tasks suffer from a length bias, suggesting benchmarks should balance tasks with varying output lengths. Second, we demonstrate that factor analysis successfully identifies reasonable yet surprising VL skill factors, suggesting benchmarks could leverage similar analyses for task selection. Finally, we present a new dataset, OLIVE (https://github.com/jq-zh/olive-dataset), which simulates user instructions in the wild and presents challenges dissimilar to all datasets we tested. Our findings contribute to the design of balanced and broad-coverage vision-language evaluation methods.

Purpose: Large Language Models (LLMs) hold significant promise for medical applications. Retrieval Augmented Generation (RAG) emerges as a promising approach for customizing domain knowledge in LLMs. This case study presents the development and evaluation of an LLM-RAG pipeline tailored for healthcare, focusing specifically on preoperative medicine. Methods: We developed an LLM-RAG model using 35 preoperative guidelines and tested it against human-generated responses, with a total of 1260 responses evaluated. The RAG process involved converting clinical documents into text using Python-based frameworks like LangChain and Llamaindex, and processing these texts into chunks for embedding and retrieval. Vector storage techniques and selected embedding models to optimize data retrieval, using Pinecone for vector storage with a dimensionality of 1536 and cosine similarity for loss metrics. Human-generated answers, provided by junior doctors, were used as a comparison. Results: The LLM-RAG model generated answers within an average of 15-20 seconds, significantly faster than the 10 minutes typically required by humans. Among the basic LLMs, GPT4.0 exhibited the best accuracy of 80.1%. This accuracy was further increased to 91.4% when the model was enhanced with RAG. Compared to the human-generated instructions, which had an accuracy of 86.3%, the performance of the GPT4.0 RAG model demonstrated non-inferiority (p=0.610). Conclusions: In this case study, we demonstrated a LLM-RAG model for healthcare implementation. The pipeline shows the advantages of grounded knowledge, upgradability, and scalability as important aspects of healthcare LLM deployment., Comment: NA

Immediate feedback has been shown to improve student learning. In programming courses, immediate, automated feedback is typically provided in the form of pre-defined test cases run by a submission platform. While these are excellent for highlighting the presence of logical errors, they do not provide novice programmers enough scaffolding to help them identify where an error is or how to fix it. To address this, several tools have been developed that provide richer feedback in the form of program repairs. Studies of such tools, however, tend to focus more on whether correct repairs can be generated, rather than how novices are using them. In this paper, we describe our experience of using CLARA, an automated repair tool, to provide feedback to novices. First, we extended CLARA to support a larger subset of the Python language, before integrating it with the Jupyter Notebooks used for our programming exercises. Second, we devised a preliminary study in which students tackled programming problems with and without support of the tool using the 'think aloud' protocol. We found that novices often struggled to understand the proposed repairs, echoing the well-known challenge to understand compiler/interpreter messages. Furthermore, we found that students valued being told where a fix was needed - without necessarily the fix itself - suggesting that 'less may be more' from a pedagogical perspective., Comment: Experience report accepted by the International Conference on Teaching, Assessment, and Learning for Engineering (TALE'23)

The optimization of nanomaterial synthesis using numerous synthetic variables is considered to be extremely laborious task because the conventional combinatorial explorations are prohibitively expensive. In this work, we report an autonomous experimentation platform developed for the bespoke design of nanoparticles (NPs) with targeted optical properties. This platform operates in a closed-loop manner between a batch synthesis module of NPs and a UV- Vis spectroscopy module, based on the feedback of the AI optimization modeling. With silver (Ag) NPs as a representative example, we demonstrate that the Bayesian optimizer implemented with the early stopping criterion can efficiently produce Ag NPs precisely possessing the desired absorption spectra within only 200 iterations (when optimizing among five synthetic reagents). In addition to the outstanding material developmental efficiency, the analysis of synthetic variables further reveals a novel chemistry involving the effects of citrate in Ag NP synthesis. The amount of citrate is a key to controlling the competitions between spherical and plate-shaped NPs and, as a result, affects the shapes of the absorption spectra as well. Our study highlights both capabilities of the platform to enhance search efficiencies and to provide a novel chemical knowledge by analyzing datasets accumulated from the autonomous experimentations.

This research paper presents the findings of two experimental studies that explore the use of ChatGPT as a tool for theory prototyping. The objective of the studies is to assess ChatGPT's ability to comprehend theoretical concepts and differentiate between constructs. During the experiments, duplicated responses were identified in both Study 1 and Study 2, with duplicate response rates of 26.25% and 40% respectively. The results of the experiments indicate that ChatGPT can generate responses aligned with the constructs of the Technology Acceptance Model (TAM). The loading and reliability coefficients demonstrate the validity of the models, with Study 1 achieving an R-squared value of 82% and Study 2 achieving 71%. In Study 2, two items with negative wording exhibited low loadings and were subsequently removed from the model. Both studies exhibit reasonable discriminant validity despite high correlations among the TAM constructs. The experiments reveal potential biases in the generated samples, particularly regarding gender and usage experiences. These biases may impact the responses of constructs and should be considered when interpreting ChatGPT's conceptual capabilities. In sum, ChatGPT shows promise as a tool for theory prototyping, generating relevant responses aligned with theoretical constructs. However, further investigation is needed to address limitations such as duplicated responses, variations in prompts, and the generalizability of findings to different contexts., Comment: 15 pages, 6 figures, 12 tables

Large-scale pre-training and instruction tuning have been successful at creating general-purpose language models with broad competence. However, building general-purpose vision-language models is challenging due to the rich input distributions and task diversity resulting from the additional visual input. Although vision-language pretraining has been widely studied, vision-language instruction tuning remains under-explored. In this paper, we conduct a systematic and comprehensive study on vision-language instruction tuning based on the pretrained BLIP-2 models. We gather 26 publicly available datasets, covering a wide variety of tasks and capabilities, and transform them into instruction tuning format. Additionally, we introduce an instruction-aware Query Transformer, which extracts informative features tailored to the given instruction. Trained on 13 held-in datasets, InstructBLIP attains state-of-the-art zero-shot performance across all 13 held-out datasets, substantially outperforming BLIP-2 and larger Flamingo models. Our models also lead to state-of-the-art performance when finetuned on individual downstream tasks (e.g., 90.7% accuracy on ScienceQA questions with image contexts). Furthermore, we qualitatively demonstrate the advantages of InstructBLIP over concurrent multimodal models. All InstructBLIP models are open-sourced at https://github.com/salesforce/LAVIS/tree/main/projects/instructblip., Comment: preprint

Very low-resolution face recognition (VLRFR) poses unique challenges, such as tiny regions of interest and poor resolution due to extreme standoff distance or wide viewing angle of the acquisition devices. In this paper, we study principled approaches to elevate the recognizability of a face in the embedding space instead of the visual quality. We first formulate a robust learning-based face recognizability measure, namely recognizability index (RI), based on two criteria: (i) proximity of each face embedding against the unrecognizable faces cluster center and (ii) closeness of each face embedding against its positive and negative class prototypes. We then devise an index diversion loss to push the hard-to-recognize face embedding with low RI away from unrecognizable faces cluster to boost the RI, which reflects better recognizability. Additionally, a perceptibility attention mechanism is introduced to attend to the most recognizable face regions, which offers better explanatory and discriminative traits for embedding learning. Our proposed model is trained end-to-end and simultaneously serves recognizability-aware embedding learning and face quality estimation. To address VLRFR, our extensive evaluations on three challenging low-resolution datasets and face quality assessment demonstrate the superiority of the proposed model over the state-of-the-art methods., Comment: Accepted to CVPR23

Large language models (LLMs) have demonstrated excellent zero-shot generalization to new language tasks. However, effective utilization of LLMs for zero-shot visual question-answering (VQA) remains challenging, primarily due to the modality disconnection and task disconnection between LLM and VQA task. End-to-end training on vision and language data may bridge the disconnections, but is inflexible and computationally expensive. To address this issue, we propose \emph{Img2Prompt}, a plug-and-play module that provides the prompts that can bridge the aforementioned modality and task disconnections, so that LLMs can perform zero-shot VQA tasks without end-to-end training. In order to provide such prompts, we further employ LLM-agnostic models to provide prompts that can describe image content and self-constructed question-answer pairs, which can effectively guide LLM to perform zero-shot VQA tasks. Img2Prompt offers the following benefits: 1) It can flexibly work with various LLMs to perform VQA. 2)~Without the needing of end-to-end training, it significantly reduces the cost of deploying LLM for zero-shot VQA tasks. 3) It achieves comparable or better performance than methods relying on end-to-end training. For example, we outperform Flamingo \cite{Deepmind:Flamingo2022} by 5.6\% on VQAv2. On the challenging A-OKVQA dataset, our method even outperforms few-shot methods by as much as 20\%., Comment: CVPR 2023 Camera Ready Version

Visual question answering (VQA) is a hallmark of vision and language reasoning and a challenging task under the zero-shot setting. We propose Plug-and-Play VQA (PNP-VQA), a modular framework for zero-shot VQA. In contrast to most existing works, which require substantial adaptation of pretrained language models (PLMs) for the vision modality, PNP-VQA requires no additional training of the PLMs. Instead, we propose to use natural language and network interpretation as an intermediate representation that glues pretrained models together. We first generate question-guided informative image captions, and pass the captions to a PLM as context for question answering. Surpassing end-to-end trained baselines, PNP-VQA achieves state-of-the-art results on zero-shot VQAv2 and GQA. With 11B parameters, it outperforms the 80B-parameter Flamingo model by 8.5% on VQAv2. With 738M PLM parameters, PNP-VQA achieves an improvement of 9.1% on GQA over FewVLM with 740M PLM parameters. Code is released at https://github.com/salesforce/LAVIS/tree/main/projects/pnp-vqa, Comment: EMNLP 2022 (Findings); correct typos in Equation 2 on page 4

Although robot-based automation in chemistry laboratories can accelerate the material development process, surveillance-free environments may lead to dangerous accidents primarily due to machine control errors. Object detection techniques can play vital roles in addressing these safety issues; however, state-of-the-art detectors, including single-shot detector (SSD) models, suffer from insufficient accuracy in environments involving complex and noisy scenes. With the aim of improving safety in a surveillance-free laboratory, we report a novel deep learning (DL)-based object detector, namely, DenseSSD. For the foremost and frequent problem of detecting vial positions, DenseSSD achieved a mean average precision (mAP) over 95% based on a complex dataset involving both empty and solution-filled vials, greatly exceeding those of conventional detectors; such high precision is critical to minimizing failure-induced accidents. Additionally, DenseSSD was observed to be highly insensitive to the environmental changes, maintaining its high precision under the variations of solution colors or testing view angles. The robustness of DenseSSD would allow the utilized equipment settings to be more flexible. This work demonstrates that DenseSSD is useful for enhancing safety in an automated material synthesis environment, and it can be extended to various applications where high detection accuracy and speed are both needed.

Recently, the transformer model has been successfully employed for the multi-view 3D reconstruction problem. However, challenges remain on designing an attention mechanism to explore the multiview features and exploit their relations for reinforcing the encoding-decoding modules. This paper proposes a new model, namely 3D coarse-to-fine transformer (3D-C2FT), by introducing a novel coarse-to-fine(C2F) attention mechanism for encoding multi-view features and rectifying defective 3D objects. C2F attention mechanism enables the model to learn multi-view information flow and synthesize 3D surface correction in a coarse to fine-grained manner. The proposed model is evaluated by ShapeNet and Multi-view Real-life datasets. Experimental results show that 3D-C2FT achieves notable results and outperforms several competing models on these datasets., Comment: Accepted by Asian Conference on Computer Vision (ACCV) 2022

Accurate predictions of the failure progression of structural materials is critical for preventing failure-induced accidents. Despite considerable mechanics modeling-based efforts, accurate prediction remains a challenging task in real-world environments due to unexpected damage factors and defect evolutions. Here, we report a novel method for predicting material failure characteristics that uniquely combines nondestructive X-ray computed tomography (X-CT), persistent homology (PH), and deep multimodal learning (DML). The combined method exploits the microstructural defect state at the time of material examination as an input, and outputs the failure-related properties. Our method is demonstrated to be effective using two types of fracture datasets (tensile and fatigue datasets) with ferritic low alloy steel as a representative structural material. The method achieves a mean absolute error (MAE) of 0.09 in predicting the local strain with the tensile dataset and an MAE of 0.14 in predicting the fracture progress with the fatigue dataset. These high accuracies are mainly due to PH processing of the X-CT images, which transforms complex and noisy three-dimensional X-CT images into compact two-dimensional persistence diagrams that preserve key topological features such as the internal void size, density, and distribution. The combined PH and DML processing of 3D X-CT data is our unique approach enabling reliable failure predictions at the time of material examination based on void topology progressions, and the method can be extended to various nondestructive failure tests for practical use.

Recent advances in deep learning and computer vision offer an excellent opportunity to investigate high-level visual analysis tasks such as human localization and human pose estimation. Although the performance of human localization and human pose estimation has significantly improved in recent reports, they are not perfect and erroneous localization and pose estimation can be expected among video frames. Studies on the integration of these techniques into a generic pipeline that is robust to noise introduced from those errors are still lacking. This paper fills the missing study. We explored and developed two working pipelines that suited the visual-based positioning and pose estimation tasks. Analyses of the proposed pipelines were conducted on a badminton game. We showed that the concept of tracking by detection could work well, and errors in position and pose could be effectively handled by a linear interpolation technique using information from nearby frames. The results showed that the Visual-based Positioning and Pose Estimation could deliver position and pose estimations with good spatial and temporal resolutions., Comment: This paper is the expanded version of our paper titled Visual-based Positioning and Pose Estimation, ICONIP (4) 2020: 410-417

Deep neural networks with lower precision weights and operations at inference time have advantages in terms of the cost of memory space and accelerator power. The main challenge associated with the quantization algorithm is maintaining accuracy at low bit-widths. We propose learned gradient linear symmetric quantization (LG-LSQ) as a method for quantizing weights and activation functions to low bit-widths with high accuracy in integer neural network processors. First, we introduce the scaling simulated gradient (SSG) method for determining the appropriate gradient for the scaling factor of the linear quantizer during the training process. Second, we introduce the arctangent soft round (ASR) method, which differs from the straight-through estimator (STE) method in its ability to prevent the gradient from becoming zero, thereby solving the discrete problem caused by the rounding process. Finally, to bridge the gap between full-precision and low-bit quantization networks, we propose the minimize discretization error (MDE) method to determine an accurate gradient in backpropagation. The ASR+MDE method is a simple alternative to the STE method and is practical for use in different uniform quantization methods. In our evaluation, the proposed quantizer achieved full-precision baseline accuracy in various 3-bit networks, including ResNet18, ResNet34, and ResNet50, and an accuracy drop of less than 1% in the quantization of 4-bit weights and 4-bit activations in lightweight models such as MobileNetV2 and ShuffleNetV2.

With the recent prevalence of remote education, academic assessments are often conducted online, leading to further concerns surrounding assessment misconducts. This paper investigates the potentials of online assessment misconduct (e-cheating) and proposes practical countermeasures against them. The mechanism for detecting the practices of online cheating is presented in the form of an e-cheating intelligent agent, comprising of an internet protocol (IP) detector and a behavioural monitor. The IP detector is an auxiliary detector which assigns randomised and unique assessment sets as an early procedure to reduce potential misconducts. The behavioural monitor scans for irregularities in assessment responses from the candidates, further reducing any misconduct attempts. This is highlighted through the proposal of the DenseLSTM using a deep learning approach. Additionally, a new PT Behavioural Database is presented and made publicly available. Experiments conducted on this dataset confirm the effectiveness of the DenseLSTM, resulting in classification accuracies of up to 90.7%.

The inner operations of the human brain as a biological processing system remain largely a mystery. Inspired by the function of the human brain and based on the analysis of simple neural network systems in other species, such as Drosophila, neuromorphic computing systems have attracted considerable interest. In cellular-level connectomics research, we can identify the characteristics of biological neural network, called population, which constitute not only recurrent fullyconnection in network, also an external-stimulus and selfconnection in each neuron. Relying on low data bandwidth of spike transmission in network and input data, Spiking Neural Networks exhibit low-latency and low-power design. In this study, we proposed a configurable population-based digital spiking neuromorphic processor in 180nm process technology with two configurable hierarchy populations. Also, these neurons in the processor can be configured as novel models, integer quadratic integrate-and-fire neuron models, which contain an unsigned 8-bit membrane potential value. The processor can implement intelligent decision making for avoidance in real-time. Moreover, the proposed approach enables the developments of biomimetic neuromorphic system and various low-power, and low-latency inference processing applications.

In vision domain, large-scale natural datasets typically exhibit long-tailed distribution which has large class imbalance between head and tail classes. This distribution poses difficulty in learning good representations for tail classes. Recent developments have shown good long-tailed model can be learnt by decoupling the training into representation learning and classifier balancing. However, these works pay insufficient consideration on the long-tailed effect on representation learning. In this work, we propose interpolative centroid contrastive learning (ICCL) to improve long-tailed representation learning. ICCL interpolates two images from a class-agnostic sampler and a class-aware sampler, and trains the model such that the representation of the interpolative image can be used to retrieve the centroids for both source classes. We demonstrate the effectiveness of our approach on multiple long-tailed image classification benchmarks. Our result shows a significant accuracy gain of 2.8% on the iNaturalist 2018 dataset with a real-world long-tailed distribution., Comment: Add in acknowledgment

Novice programmers often struggle with problem solving due to the high cognitive loads they face. Furthermore, many introductory programming courses do not explicitly teach it, assuming that problem solving skills are acquired along the way. In this paper, we present 'PCDIT', a non-linear problem solving framework that provides scaffolding to guide novice programmers through the process of transforming a problem specification into an implemented and tested solution for an imperative programming language. A key distinction of PCDIT is its focus on developing concrete cases for the problem early without actually writing test code: students are instead encouraged to think about the abstract steps from inputs to outputs before mapping anything down to syntax. We reflect on our experience of teaching an introductory programming course using PCDIT, and report the results of a survey that suggests it helped students to break down challenging problems, organise their thoughts, and reach working solutions., Comment: Accepted by the 34th Conference on Software Engineering Education and Training (CSEE&T 2022): Special Track of the 55th Hawaii International Conference on System Sciences (HICSS 2022)

By combining IFS with ExAO we are now able to resolve objects close to the diffraction-limit of large telescopes, exploring new science cases. We introduce an IFU designed to couple light with a minimal platescale from the SCExAO facility at NIR wavelengths to a SM spectrograph. The IFU has a 3D-printed MLA on top of a custom SM MCF, to optimize the coupling of light into the fiber cores. We demonstrate the potential of the instrument via initial results from the first on-sky runs at the 8.2 m Subaru Telescope with a spectrograph using off-the-shelf optics, allowing for rapid development with low cost.

This study addresses the current issues in online assessments, which are particularly relevant during the Covid-19 pandemic. Our focus is on academic dishonesty associated with online assessments. We investigated the prevalence of potential e-cheating using a case study and propose preventive measures that could be implemented. We have utilised an e-cheating intelligence agent as a mechanism for detecting the practices of online cheating, which is composed of two major modules: the internet protocol (IP) detector and the behaviour detector. The intelligence agent monitors the behaviour of the students and has the ability to prevent and detect any malicious practices. It can be used to assign randomised multiple-choice questions in a course examination and be integrated with online learning programs to monitor the behaviour of the students. The proposed method was tested on various data sets confirming its effectiveness. The results revealed accuracies of 68% for the deep neural network (DNN); 92% for the long-short term memory (LSTM); 95% for the DenseLSTM; and, 86% for the recurrent neural network (RNN).

In the new era of Extremely Large Telescopes (ELTs) currently under construction, challenging requirements drive spectrograph designs towards techniques that efficiently use a facility's light collection power. Operating in the single-mode (SM) regime, close to the diffraction limit, reduces the footprint of the instrument compared to a conventional high-resolving power spectrograph. The custom built injection fiber system with 3D-printed micro-lenses on top of it for the replicable high-resolution exoplanet and asteroseismology spectrograph at Subaru in combination with extreme adaptive optics of SCExAO, proved its high efficiency in a lab environment, manifesting up to ~77% of the theoretical predicted performance.

Periocular biometric, the peripheral area of the ocular, is a collaborative alternative to the face, especially when the face is occluded or masked. However, in practice, sole periocular biometric capture the least salient facial features, thereby lacking discriminative information, particularly in wild environments. To address these problems, we transfer discriminatory information from the face to support the training of a periocular network by using knowledge distillation. Specifically, we leverage face images for periocular embedding learning, but periocular alone is utilized for identity identification or verification. To enhance periocular embeddings by face effectively, we proposeConsistent Knowledge Distillation (CKD) that imposes consistency between face and periocular networks across prediction and feature layers. We find that imposing consistency at the prediction layer enables (1) extraction of global discriminative relationship information from face images and (2) effective transfer of the information from the face network to the periocular network. Particularly, consistency regularizes the prediction units to extract and store profound inter-class relationship information of face images. (3) The feature layer consistency, on the other hand, makes the periocular features robust against identity-irrelevant attributes. Overall, CKD empowers the sole periocular network to produce robust discriminative embeddings for periocular recognition in the wild. We theoretically and empirically validate the core principles of the distillation mechanism in CKD, discovering that CKD is equivalent to label smoothing with a novel sparsity-oriented regularizer that helps the network prediction to capture the global discriminative relationship. Extensive experiments reveal that CKD achieves state-of-the-art results on standard periocular recognition benchmark datasets., Comment: Accepted to Neurocomputing

Convolutional neural networks (CNNs) play a key role in deep learning applications. However, the large storage overheads and the substantial computation cost of CNNs are problematic in hardware accelerators. Computing-in-memory (CIM) architecture has demonstrated great potential to effectively compute large-scale matrix-vector multiplication. However, the intensive multiply and accumulation (MAC) operations executed at the crossbar array and the limited capacity of CIM macros remain bottlenecks for further improvement of energy efficiency and throughput. To reduce computation costs, network pruning and quantization are two widely studied compression methods to shrink the model size. However, most of the model compression algorithms can only be implemented in digital-based CNN accelerators. For implementation in a static random access memory (SRAM) CIM-based accelerator, the model compression algorithm must consider the hardware limitations of CIM macros, such as the number of word lines and bit lines that can be turned on at the same time, as well as how to map the weight to the SRAM CIM macro. In this study, a software and hardware co-design approach is proposed to design an SRAM CIM-based CNN accelerator and an SRAM CIM-aware model compression algorithm. To lessen the high-precision MAC required by batch normalization (BN), a quantization algorithm that can fuse BN into the weights is proposed. Furthermore, to reduce the number of network parameters, a sparsity algorithm that considers a CIM architecture is proposed. Last, MARS, a CIM-based CNN accelerator that can utilize multiple SRAM CIM macros as processing units and support a sparsity neural network, is proposed., Comment: IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems 2021

We develop a single-layer waveguide surface grating structure to vertically couple near infrared (NIR) light at ~1.55 um wavelength from a large area (~100 um length scale) Si waveguide on a Silicon-On-Insulator (SOI) substrates to free-space for high-power laser applications. Our design approach is based on the optimization of local emission angles and the out-coupling intensities. Simulation results show that a focal spot with a 1/e2 width of 3.82 um can be achieved at the desired focal position, with 33% (-4.81 dB) simulated source to free-space focusing efficiency, while initial measurements show an efficiency of 22% (-6.58 dB).

Despite the significant progress made in studying the lightning phenomenon, precise location and mapping of its occurrence remain a challenge. Lightning mapping can be determined by studying the electromagnetic radiation accompanying the lightning discharges. It can contribute substantially to efforts made to protect lives and valuable assets. There are three main methods used to locate lightning, which are Magnetic Direction Finder (MDF), Time of Arrival (TOA), and Interferometer (ITF). A thorough study of these methods provides researchers with a guide to better understand and progress in this field. This paper reviews existing approaches used to locate and map lightning within these three methods. We study the implemented techniques, analyze their merits and demerits, and sort them in a way that facilitates extracting opportunities for further improvements. We conclude that for better development in determining the location and map of lightning, improving the processing of lightning signals and filtering the associated noise with it is essential. This includes introducing new processing methods such as wavelet transformation instead of the traditional cross-correlation. The use of artificial intelligence may also contribute a lot, particularly deep learning, to determining the type of lightning, which enables better mapping for the lightning and its occurrence. We also could conclude that unlike MDF and TOA, which can locate the lightning strike points, ITF can produce lightning discharge propagation images that can unveil the mechanism of lightning discharges. Finally, this paper serves as a reference for researchers focusing on lightning mapping to give them insight into the field., Comment: 24 pages, 5 figures

The robust and automated determination of crystal symmetry is of utmost importance in material characterization and analysis. Recent studies have shown that deep learning (DL) methods can effectively reveal the correlations between X-ray or electron-beam diffraction patterns and crystal symmetry. Despite their promise, most of these studies have been limited to identifying relatively few classes into which a target material may be grouped. On the other hand, the DL-based identification of crystal symmetry suffers from a drastic drop in accuracy for problems involving classification into tens or hundreds of symmetry classes (e.g., up to 230 space groups), severely limiting its practical usage. Here, we demonstrate that a combined approach of shaping diffraction patterns and implementing them in a multistream DenseNet (MSDN) substantially improves the accuracy of classification. Even with an imbalanced dataset of 108,658 individual crystals sampled from 72 space groups, our model achieves 80.2% space group classification accuracy, outperforming conventional benchmark models by 17-27 percentage points (%p). The enhancement can be largely attributed to the pattern shaping strategy, through which the subtle changes in patterns between symmetrically close crystal systems (e.g., monoclinic vs. orthorhombic or trigonal vs. hexagonal) are well differentiated. We additionally find that the novel MSDN architecture is advantageous for capturing patterns in a richer but less redundant manner relative to conventional convolutional neural networks. The newly proposed protocols in regard to both input descriptor processing and DL architecture enable accurate space group classification and thus improve the practical usage of the DL approach in crystal symmetry identification.

The integration of large-scale wind farms and large-scale charging stations for electric vehicles with the electricity grids necessitate energy storage support for both technologies. Matching the energy variability of the wind farms with the demand variability of the electric vehicles (EVs) off-grid could potentially eliminate the need for expensive energy storage technologies required to stabilize the grid. The objective of this paper is to investigate the feasibility of using wind generation as direct energy source to power the EV charging stations. An interval-based approach corresponding to the time slot of EV charging is introduced for wind energy conversion and analyzed using different constrains and criteria including, wind speed averaging time interval, various turbines manufacturers, and standard high-resolution wind speed data sets. We performed a piecewise recursive of wind turbines' output energy to measure the EV charging efficiency. Wind averaging results show that the three minutes intervals have increased the total number of EVs by more than 80% compared to one-- and two -minute intervals. The potential cost reduction due to decoupling of both technologies from the utility grid, energy storage systems, and the associated energy conversion power electronics has merit and research in this direction is worth pursuing, Comment: 14 pages, 10 figures

Wind energy has gained a huge interest in the recent years in various countries due to the high demand of energy and the shortage of traditional electricity sources. This is because it is cost effective and environmentally friendly source that could contribute significantly to the reduction of the ever-increasing carbon emissions. Wind energy is one of the fastest growing green technology worldwide with a total generation share of 564 GW as the end of 2018. In Malaysia, wind energy has been a topic of interest in both academia and green energy industry. In this paper, the current status of wind energy research in Malaysia is reviewed. Different contributing factors such as potentiality and assessments, wind speed and direction modelling, wind prediction and spatial mapping, and optimal sizing of wind farms are extensively discussed. This paper discusses the progress of all wind studies and presents conclusions and recommendations to improve wind research in Malaysia., Comment: 25 pages, 3 figures

Traditional pen and paper exams are inadequate for modern university programming courses as they are misaligned with pedagogies and learning objectives that target practical coding ability. Unfortunately, many institutions lack the resources or space to be able to run assessments in dedicated computer labs. This has motivated the development of bring-your-own-device (BYOD) exam formats, allowing students to program in a similar environment to how they learnt, but presenting instructors with significant additional challenges in preventing plagiarism and cheating. In this paper, we describe a BYOD exam solution based on lockdown browsers, software which temporarily turns students' laptops into secure workstations with limited system or internet access. We combine the use of this technology with a learning management system and cloud-based programming tool to facilitate conceptual and practical programming questions that can be tackled in an interactive but controlled environment. We reflect on our experience of implementing this solution for a major undergraduate programming course, highlighting our principal lesson that policies and support mechanisms are as important to consider as the technology itself., Comment: Accepted by SIGCSE 2020

Robotic minimally invasive interventions typically require using more than two instruments. We thus developed a foot pedal interface which allows the user to control a robotic arm (simultaneously to working with the hands) with four degrees of freedom in continuous directions and speeds. This paper evaluates and compares the performances of ten naive operators in using this new pedal interface and a traditional button interface in completing tasks. These tasks are geometrically complex path-following tasks similar to those in laparoscopic training, and the traditional button interface allows axis-by-axis control with constant speeds. Precision, time, and smoothness of the subjects' control movements for these tasks are analysed. The results demonstrate that the pedal interface can be used to control a robot for complex motion tasks. The subjects kept the average error rate at a low level of around 2.6% with both interfaces, but the pedal interface resulted in about 30% faster operation speed and 60% smoother movement, which indicates improved efficiency and user experience as compared with the button interface. The results of a questionnaire show that the operators found that controlling the robot with the pedal interface was more intuitive, comfortable, and less tiring than using the button interface., Comment: 7 pages, submit to 2019 IROS RA-Letter

In spite of the advancements made in the periocular recognition, the dataset and periocular recognition in the wild remains a challenge. In this paper, we propose a multilayer fusion approach by means of a pair of shared parameters (dual-stream) convolutional neural network where each network accepts RGB data and a novel colour-based texture descriptor, namely Orthogonal Combination-Local Binary Coded Pattern (OC-LBCP) for periocular recognition in the wild. Specifically, two distinct late-fusion layers are introduced in the dual-stream network to aggregate the RGB data and OC-LBCP. Thus, the network beneficial from this new feature of the late-fusion layers for accuracy performance gain. We also introduce and share a new dataset for periocular in the wild, namely Ethnic-ocular dataset for benchmarking. The proposed network has also been assessed on one publicly available dataset, namely UBIPr. The proposed network outperforms several competing approaches on these datasets., Comment: Accepted in International Conference On Biometrics 2019

In robotic surgery, the surgeon controls robotic instruments using dedicated interfaces. One critical limitation of current interfaces is that they are designed to be operated by only the hands. This means that the surgeon can only control at most two robotic instruments at one time while many interventions require three instruments. This paper introduces a novel four-degree-of-freedom foot-machine interface which allows the surgeon to control a third robotic instrument using the foot, giving the surgeon a "third hand". This interface is essentially a parallel-serial hybrid mechanism with springs and force sensors. Unlike existing switch-based interfaces that can only un-intuitively generate motion in discrete directions, this interface allows intuitive control of a slave robotic arm in continuous directions and speeds, naturally matching the foot movements with dynamic force & position feedbacks. An experiment with ten naive subjects was conducted to test the system. In view of the significant variance of motion patterns between subjects, a subject-specific mapping from foot movements to command outputs was developed using Independent Component Analysis (ICA). Results showed that the ICA method could accurately identify subjects' foot motion patterns and significantly improve the prediction accuracy of motion directions from 68% to 88% as compared with the forward kinematics-based approach. This foot-machine interface can be applied for the teleoperation of industrial/surgical robots independently or in coordination with hands in the future., Comment: 11 pages,10 figures, submit to the journal of IEEE/ASME Transactions on Mechatronics with the status of under review

A nonlinear Schr\"{o}dinger (NLS) equation with an effect of viscosity is derived from a Korteweg-de Vries (KdV) equation modified with viscosity using the method of multiple time-scales. It is well known that the plane-wave solution of the NLS equation exhibits modulational instability phenomenon. In this paper, the modulational instability of the plane-wave solution of the NLS equation modified with viscosity is investigated. The corresponding modulational dispersion relation is expressed as a quadratic equation with complex-valued coefficients. By restricting the modulational wavenumber into the case of narrow-banded spectra, it is observed that a type of dissipation, in this case, the effect of viscosity, stabilizes the modulational instability, as confirmed by earlier findings., Comment: 11 pages, 3 figures

The $g$-theorem is a momentous result in combinatorics that gives a complete numerical characterization of the face numbers of simplicial convex polytopes. The $g$-conjecture asserts that the same numerical conditions given in the $g$-theorem also characterizes the face numbers of all simplicial spheres, or even more generally, all simplicial homology spheres. In this paper, we prove the $g$-conjecture for simplicial $\mathbb{R}$-homology spheres. A key idea in our proof is a new algebra structure for polytopal complexes. Given a polytopal $d$-complex $\Delta$, we use ideas from rigidity theory to construct a graded Artinian $\mathbb{R}$-algebra $\Psi(\Delta,\nu)$ of stresses on a PL realization $\nu$ of $\Delta$ in $\mathbb{R}^d$, where overlapping realized $d$-faces are allowed. In particular, we prove that if $\Delta$ is a simplicial $\mathbb{R}$-homology sphere, then for generic PL realizations $\nu$, the stress algebra $\Psi(\Delta,\nu)$ is Gorenstein and has the weak Lefschetz property., Comment: The multiplication of stresses in Thm. 5.2 is not well-defined. We have a corrected multiplication map, which introduces a coefficient that is no longer always 1 for each summand. However, subsequent proof approaches for Sec. 8-11 require this (incorrect) coefficient 1 for each summand, which we do not know how to fix. Thus, our proof approaches for all main results do not work

We report a new computational model for simulations of electromagnetic interactions with semiconductor quantum well(s) (SQW) in complex electromagnetic geometries using the finite difference time domain (FDTD) method. The presented model is based on an approach of spanning a large number of electron transverse momentum states in each SQW sub-band (multi-band) with a small number of discrete multi-electron states (multi-level, multi-electron). This enables accurate and efficient two dimensional (2-D) and 3-D simulations of nanophotonic devices with SQW active media. The model includes the following features: (1) Optically induced interband transitions between various SQW conduction and heavy-hole or light-hole sub-bands are considered. (2) Novel intra sub-band and inter sub-band transition terms are derived to thermalize the electron and hole occupational distributions to the correct Fermi-Dirac distributions. (3) The terms in (2) result in an explicit update scheme which circumvents numerically cumbersome iterative procedures. This significantly augments computational efficiency. (4) Explicit update terms to account for carrier leakage to unconfined states are derived which thermalize the bulk and SQW populations to a common quasi-equilibrium Fermi-Dirac distribution. (5) Auger recombination and intervalence band absorption are included. The model is validated by comparisons to analytic band filling calculations, simulations of SQW optical gain spectra and photonic crystal lasers.

The number of ethnic Chinese students in schools across Australian cities is small but increasing. It is important to understand how these students socialise into the Australian (mathematics) education system, so that we can better facilitate their education experiences in ways which optimise their potential to learn. This paper reports on part of a larger study which seeks to deepen our knowledge in this area. The research question addressed by this paper is: what are the preferences amongst three types of mathematical tasks of Grade 5 and 6 students from Chongqing, China? Through the administration of a questionnaire to 1109 students, it was found that across the topics of Number and Geometry, contextualised tasks were the most preferred by Chinese students. "Challenging", "easy to do" and "involving a model" were students' reasons for preferring particular task types. The significance of providing all students with a diverse range of task types, at the same time providing them with opportunities to be challenged and to experience success, are emphasized.

This paper reports on a qualitative research study exploring the socialisation experiences of immigrant secondary mathematics teachers practising in Australia. Teacher perception of differences in the ways their respective home and the Australian (host) cultures value aspects of mathematics teaching and learning was observed to lead to dissonance. Their negotiation about these differences highlighted the role played by personally-held values. Although each teacher participant adopts different approaches to mediate the different perceived value differences, the inclusive approaches of amalgamation and appropriation were most widely adopted. Implications towards optimising mathematics pedagogy, and towards meaningful professional support for mathematics teachers in transition, are suggested. (Contains 1 table.) [For complete proceedings, see ED496851.]

Simulation of detecting photoelectrons using multi-level multi-electron (MLME) finite-difference time-domain (FDTD) method with an application to near-field subwavelength imaging based on semiconductor nanophotodetector (NPD) array is reported. The photocurrents from the photodiode pixels are obtained to explore the resolution of this novel NPD device for subwavelength imaging. One limiting factor of the NPD device is the optical power coupling between adjacent detector pixels. We investigate such power coupling in the presence of absorbing media as well as the spatial distributions of the electric field and photoelectron density using the MLME FDTD simulation. Our results show that the detection resolution is about one tenth of the operating wavelength, which is comparable to that of a near-field scanning optical microscope based on metal clad tapered fiber., Comment: 7 pages, 4 figures

Values are taught in every lesson; however, in mathematics classes this seems to be implicit rather than explicit. This paper discusses an approach to explore the values that teachers teach in mathematics sessions and outlines methodological issues encountered in researching these values. First, ways of researching values is discussed and worked with individual teachers using a cycle of a preliminary interview, a classroom observation, and a post-observation debriefing interview on the same day. Overall the methods chosen for the exploration seem to be useful in collecting what appears to be rich data for insights to be gained as to what values teachers can plan to teach and what values remain implicit in their teaching. One crucial area that has emerged is finding a common language with which meaningful dialogue can occur. One of the issues that remains is to understand more fully the language that teachers use as they try to share with researchers and each other what values they wish to teach. (Contains 14 references.) (ASK)