Abstract

This paper studies the contextual multi-armed bandit problem with fairness and privacy guarantees in a federated setting. It proposes a collaborative algorithm, Fed-FairX-LinUCB that achieves sublinear fairness regret and can be adapted to ensure differential privacy. The key challenge is designing a communication protocol that balances privacy and regret. The proposed protocol achieves both sub-linear fairness regret and effective use of privacy budget. Experiments validates the efficacy of both Fed-FairX-LinUCB and its private counterpart, Priv-FairX-LinUCB.

Abstract

Automated Sleep stage classification using raw single channel EEG is a critical tool for sleep quality assessment and disorder diagnosis. However, modelling the complexity and variability inherent in this signal is a challenging task, limiting their practicality and effectiveness in clinical settings. To mitigate these challenges, this study presents an end-to-end deep learning (DL) model which integrates squeeze and excitation blocks within the residual network to extract features and stacked Bi-LSTM to understand complex temporal dependencies. A distinctive aspect of this study is the adaptation of GradCam for sleep staging, marking the first instance of an explainable DL model in this domain with alignment of its decision-making with sleep expert's insights. We evaluated our model on the publically available datasets (SleepEDF-20, SleepEDF-78, and SHHS), achieving Macro-F1 scores of 82.5, 78.9, and 81.9, respectively. Additionally, a novel training efficiency enhancement strategy was implemented by increasing stride size, leading to 8x faster training times with minimal impact on performance. Comparative analyses underscore our model outperforms all existing baselines, indicating its potential for clinical usage.

Abstract

Identifying correct human postures is crucial in areas like patient care in hospitals. However, the traditional vision-based methods widely used for this purpose raise privacy concerns for the subject, and the other wearable sensor-based approaches are impractical for real-world scenarios. In this paper, we propose a contactless, privacy-conscious, and memoryefficient posture classification system based on millimeter wave (mmWave) radar. This system utilizes threedimension(3D) point-cloud data captured using Texas Instrument’s IWR1843BOOST Frequency Modulated Continuous Wave (FMCW) radar module to classify the posture of the subject. Two types of datasets are extracted from this radar data: (i) image dataset derived from the isometric view of the point-cloud data, and (ii) spatial coordinates dataset also extracted from the point-cloud data. A low-computational Tiny Machine Learning (TinyML) model is employed on the datasets for efficient implementation on embedded hardware, Raspberry Pi 3 B+. The proposed model’s parameters were quantized to 8 bits (int8), which accurately classify four postures, i.e., standing, sitting, lying, and bending, with an accuracy of 98.97% for the image data. However, to make it more computationally efficient, the int8 quantized TinyML model was trained on the spatial coordinates dataset, giving an accuracy of 96.12%. This highlights the efficiency and effectiveness of our proposed lightweight model that can be deployed on edge devices for real-world applications.

Abstract

Accurately recognizing a revisited place is crucial for embodied agents to localize and navigate. This requires visual representations to be distinct, despite strong variations in camera viewpoint and scene appearance. Existing visual place recognition pipelines encode the whole image and search for matches. This poses a fundamental challenge in matching two images of the same place captured from different camera viewpoints: the similarity of what overlaps can be dominated by the dissimilarity of what does not overlap. We address this by encoding and searching for image segments instead of the whole images. We propose to use open-set image segmentation to decompose an image into `meaningful' entities (i.e., things and stuff). This enables us to create a novel image representation as a collection of multiple overlapping subgraphs connecting a segment with its neighboring segments, dubbed SuperSegment. Furthermore, to efficiently encode these SuperSegments into compact vector representations, we propose a novel factorized representation of feature aggregation. We show that retrieving these partial representations leads to significantly higher recognition recall than the typical whole image based retrieval. Our segments-based approach, dubbed SegVLAD, sets a new state-of-the-art in place recognition on a diverse selection of benchmark datasets, while being applicable to both generic and task-specialized image encoders. Finally, we demonstrate the potential of our method to revisit anything by evaluating our method on an object instance retrieval task, which bridges the two disparate areas of research: visual place recognition and object-goal navigation, through their common aim of recognizing goal objects specific to a place.

Abstract

Lupus Nephritis classification has historically relied on labor-intensive and meticulous glomerular-level labeling of renal structures in whole slide images (WSIs). However, this approach presents a formidable challenge due to its tedious and resource-intensive nature, limiting its scalability and practicality in clinical settings. In response to this challenge, our work introduces a novel methodology that utilizes only slide-level labels, eliminating the need for granular glomerular-level labeling. A comprehensive multi-stained lupus nephritis digital histopathology WSI dataset was created from the Indian population, which is the largest of its kind. LupusNet, a deep learning MIL-based model, was developed to classify LN subtypes. The results underscore its effectiveness, achieving an AUC score of 91.0%, an F1 score of 77.3%, and an accuracy of 81.1% on our dataset in distinguishing membranous and diffused classes of LN

Abstract

The histone methyltransferase ASH1L, first discovered for its role in transcription, has been shown to accelerate the removal of ultraviolet (UV) light-induced cyclobutane pyrimidine dimers (CPDs) by nucleotide excision repair. Previous reports demonstrated that CPD excision is most efficient at transcriptional regulatory elements, including enhancers, relative to other genomic sites. Therefore, we analyzed DNA damage maps in ASH1L-proficient and ASH1L-deficient cells to understand how ASH1L controls enhancer stability. This comparison showed that ASH1L protects enhancer sequences against the induction of CPDs besides stimulating repair activity. ASH1L reduces CPD formation at C–containing but not at TT dinucleotides, and no protection occurs against pyrimidine-(6,4)-pyrimidone photoproducts or cisplatin crosslinks. The diminished CPD induction extends to gene promoters but excludes retrotransposons. This guardian role against CPDs in regulatory elements is associated with the presence of H3K4me3 and H3K27ac histone marks, which are known to interact with the PHD and BRD motifs of ASH1L, respectively. Molecular dynamics simulations identified a DNA-binding AT hook of ASH1L that alters the distance and dihedral angle between neighboring C nucleotides to disfavor dimerization. The loss of this protection results in a higher frequency of C–>T transitions at enhancers of skin cancers carrying ASH1L mutations compared to ASH1L-intact counterparts.

Abstract

Chemical reaction prediction, encompassing forward synthesis and retrosynthesis, stands as a fundamental challenge in organic synthesis. A widely adopted computational approach frames synthesis prediction as a sequence-to-sequence translation task, using the common SMILES representation for molecules. Current evaluation of machine learning methods for retrosynthesis assume perfect training data, overlooking imperfections in reaction equations in popular datasets, such as missing reactants, products, other physical and practical constraints such as temperature and cost, primarily driven by a focus on the target molecule. This limitation leads to an incomplete representation of viable synthetic routes, especially when multiple sets of reactants can yield a given desired product. In response to these shortcomings, this study examines the prevailing evaluation methods and introduces comprehensive metrics designed to address imperfections in the dataset. Our novel metrics not only assess absolute accuracy by comparing predicted outputs with ground truth but also introduce a nuanced evaluation approach. We provide scores for partial correctness and compute adjusted accuracy through graph matching, acknowledging the inherent complexities of retrosynthetic pathways. Additionally, we explore the impact of small molecular augmentations while preserving chemical properties and employ similarity matching to enhance the assessment of prediction quality. We introduce SynFormer, a sequence-to-sequence model tailored for SMILES representation. It incorporates architectural enhancements to the original transformer, effectively tackling the challenges of chemical reaction prediction. SynFormer achieves a top-1 accuracy of 53.2% on the USPTO-50k dataset, demonstrating an improvement over previous state-of-the-art language models while being more efficient and eliminating the need for pre-training.

Abstract

Computational History, Trade in Mountain Societies, NLP – NER, Relation Extraction, Application of LLMs.

Abstract

The spot size variation due to mechanical jitter is calculated for the Hermite Gaussian (HG) incoherent beam combination system. It is observed that the spot size increases with an increase in mechanical jitter and higher modes HG beams experience less mechanical jitter effect compared to lower mode HG beams.

Abstract

We present the first shared task on Semantic Textual Relatedness (STR). While earlier shared tasks primarily focused on semantic similarity, we instead investigate the broader phenomenon of semantic relatedness across 14 languages: Afrikaans, Algerian Arabic, Amharic, English, Hausa, Hindi, Indonesian, Kinyarwanda, Marathi, Moroccan Arabic, Modern Standard Arabic, Punjabi, Spanish, and Telugu. These languages originate from five distinct language families and are predominantly spoken in Africa and Asia – regions characterised by the relatively limited availability of NLP resources. Each instance in the datasets is a sentence pair associated with a score that represents the degree of semantic textual relatedness between the two sentences. Participating systems were asked to rank sentence pairs by their closeness in meaning (i.e., their degree of semantic relatedness) in the 14 languages in three main tracks: (a) supervised, (b) unsupervised, and (c) crosslingual. The task attracted 163 participants. We received 70 submissions in total (across all tasks) from 51 different teams, and 38 system description papers. We report on the best-performing systems as well as the most common and the most effective approaches for the three different tracks.