The data mixture for large language model pre-training significantly impacts performance, yet how to determine an effective mixture remains unclear. We propose RegMix to automatically identify a high-performing data mixture by formulating it as a regression task. RegMix involves training a set of small models with diverse data mixtures and fitting a regression model to predict their performance given their respective mixtures. With the fitted regression model, we simulate the top-ranked mixture and use it to train a large-scale model with orders of magnitude more compute. To empirically validate RegMix, we train 512 models with 1M parameters for 1B tokens of different mixtures to fit the regression model and find the optimal mixture. Using this mixture we train a 1B parameter model for 25B tokens (i.e. 1000x larger and 25x longer) which we find performs best among 64 candidate 1B parameter models with other mixtures. Further, our method demonstrates superior performance compared to human selection and achieves results that match or surpass DoReMi, while utilizing only 10% of the compute budget. Our experiments also show that (1) Data mixtures significantly impact performance with single-task performance variations of up to 14.6%; (2) Web corpora rather than data perceived as high-quality like Wikipedia have the strongest positive correlation with downstream performance; (3) Domains interact in complex ways often contradicting common sense, thus automatic approaches like RegMix are needed; (4) Data mixture effects transcend scaling laws, and our approach captures the complexity by considering all domains together. Our code is available at https://github.com/sail-sg/regmix.

INTRODUCTION: Previous studies have applied normative modeling on a single neuroimaging modality to investigate Alzheimer Disease (AD) heterogeneity. We employed a deep learning-based multimodal normative framework to analyze individual-level variation across ATN (amyloid-tau-neurodegeneration) imaging biomarkers. METHODS: We selected cross-sectional discovery (n = 665) and replication cohorts (n = 430) with available T1-weighted MRI, amyloid and tau PET. Normative modeling estimated individual-level abnormal deviations in amyloid-positive individuals compared to amyloid-negative controls. Regional abnormality patterns were mapped at different clinical group levels to assess intra-group heterogeneity. An individual-level disease severity index (DSI) was calculated using both the spatial extent and magnitude of abnormal deviations across ATN. RESULTS: Greater intra-group heterogeneity in ATN abnormality patterns was observed in more severe clinical stages of AD. Higher DSI was associated with worse cognitive function and increased risk of disease progression. DISCUSSION: Subject-specific abnormality maps across ATN reveal the heterogeneous impact of AD on the brain.

Tensor factorizations have been widely used for the task of uncovering patterns in various domains. Often, the input is time-evolving, shifting the goal to tracking the evolution of underlying patterns instead. To adapt to this more complex setting, existing methods incorporate temporal regularization but they either have overly constrained structural requirements or lack uniqueness which is crucial for interpretation. In this paper, in order to capture the underlying evolving patterns, we introduce t(emporal)PARAFAC2 which utilizes temporal smoothness regularization on the evolving factors. We propose an algorithmic framework that employs Alternating Optimization (AO) and the Alternating Direction Method of Multipliers (ADMM) to fit the model. Furthermore, we extend the algorithmic framework to the case of partially observed data. Our numerical experiments on both simulated and real datasets demonstrate the effectiveness of the temporal smoothness regularization, in particular, in the case of data with missing entries. We also provide an extensive comparison of different approaches for handling missing data within the proposed framework.

Establishing the chronology of the Vedas has interested scholars for the last two centuries. The oldest among them is Rig-Veda which has ten Mandalas, each composed separately. In this paper, we look at deciphering plausible pointers to the internal chronology of the Mandalas, by focusing on Gods and Goddesses worshiped in different Mandalas. We apply text analysis to the Mandalas using Clustering Techniques based on Cosine Similarity. Then we represent the association of deities with Mandalas using a grid-based Social Network that is amenable to chronological analysis and demonstrates the benefits of using Social Network Analysis for the problem at hand. Further, we analyze references to rivers to arrive at additional correlations. The approach used can be deployed generically to analyze other kinds of references and mentions and arrive at more substantive inferences.

End-to-end neural speaker diarization systems are able to address the speaker diarization task while effectively handling speech overlap. This work explores the incorporation of speaker information embeddings into the end-to-end systems to enhance the speaker discriminative capabilities, while maintaining their overlap handling strengths. To achieve this, we propose several methods for incorporating these embeddings along the acoustic features. Furthermore, we delve into an analysis of the correct handling of silence frames, the window length for extracting speaker embeddings and the transformer encoder size. The effectiveness of our proposed approach is thoroughly evaluated on the CallHome dataset for the two-speaker diarization task, with results that demonstrate a significant reduction in diarization error rates achieving a relative improvement of a 10.78% compared to the baseline end-to-end model.

Acoustic Soft Tactile (AST) skin is a novel sensing technology which derives tactile information from the modulation of acoustic waves travelling through the skin's embedded acoustic channels. A generalisable data-driven calibration model maps the acoustic modulations to the corresponding tactile information in the form of contact forces with their contact locations and contact geometries. AST skin technology has been highlighted for its easy customisation. As a case study, this paper discusses the possibility of using AST skin on a custom-built robotic end effector finger for strawberry handling. The paper delves into the design, prototyping, and calibration method to sensorise the end effector finger with AST skin. A real-time force-controlled gripping experiment is conducted with the sensorised finger to handle strawberries by their peduncle. The finger could successfully grip the strawberry peduncle by maintaining a preset force of 2 N with a maximum Mean Absolute Error (MAE) of 0.31 N over multiple peduncle diameters and strawberry weight classes. Moreover, this study sets confidence in the usability of AST skin in generating real-time tactile feedback for robot manipulation tasks.

Normalization layers have recently experienced a renaissance in the deep reinforcement learning and continual learning literature, with several works highlighting diverse benefits such as improving loss landscape conditioning and combatting overestimation bias. However, normalization brings with it a subtle but important side effect: an equivalence between growth in the norm of the network parameters and decay in the effective learning rate. This becomes problematic in continual learning settings, where the resulting effective learning rate schedule may decay to near zero too quickly relative to the timescale of the learning problem. We propose to make the learning rate schedule explicit with a simple re-parameterization which we call Normalize-and-Project (NaP), which couples the insertion of normalization layers with weight projection, ensuring that the effective learning rate remains constant throughout training. This technique reveals itself as a powerful analytical tool to better understand learning rate schedules in deep reinforcement learning, and as a means of improving robustness to nonstationarity in synthetic plasticity loss benchmarks along with both the single-task and sequential variants of the Arcade Learning Environment. We also show that our approach can be easily applied to popular architectures such as ResNets and transformers while recovering and in some cases even slightly improving the performance of the base model in common stationary benchmarks.

This study proposes CGRclust, a novel combination of unsupervised twin contrastive clustering of Chaos Game Representations (CGR) of DNA sequences, with convolutional neural networks (CNNs). To the best of our knowledge, CGRclust is the first method to use unsupervised learning for image classification (herein applied to two-dimensional CGR images) for clustering datasets of DNA sequences. CGRclust overcomes the limitations of traditional sequence classification methods by leveraging unsupervised twin contrastive learning to detect distinctive sequence patterns, without requiring DNA sequence alignment or biological/taxonomic labels. CGRclust accurately clustered twenty-five diverse datasets, with sequence lengths ranging from 664 bp to 100 kbp, including mitochondrial genomes of fish, fungi, and protists, as well as viral whole genome assemblies and synthetic DNA sequences. Compared with three recent clustering methods for DNA sequences (DeLUCS, iDeLUCS, and MeShClust v3.0.), CGRclust is the only method that surpasses 81.70% accuracy across all four taxonomic levels tested for mitochondrial DNA genomes of fish. Moreover, CGRclust also consistently demonstrates superior performance across all the viral genomic datasets. The high clustering accuracy of CGRclust on these twenty-five datasets, which vary significantly in terms of sequence length, number of genomes, number of clusters, and level of taxonomy, demonstrates its robustness, scalability, and versatility.

Recent advances in microscopy have enabled the rapid generation of terabytes of image data in cell biology and biomedical research. Vision-language models (VLMs) offer a promising solution for large-scale biological image analysis, enhancing researchers' efficiency, identifying new image biomarkers, and accelerating hypothesis generation and scientific discovery. However, there is a lack of standardized, diverse, and large-scale vision-language benchmarks to evaluate VLMs' perception and cognition capabilities in biological image understanding. To address this gap, we introduce {\mu}-Bench, an expert-curated benchmark encompassing 22 biomedical tasks across various scientific disciplines (biology, pathology), microscopy modalities (electron, fluorescence, light), scales (subcellular, cellular, tissue), and organisms in both normal and abnormal states. We evaluate state-of-the-art biomedical, pathology, and general VLMs on {\mu}-Bench and find that: i) current models struggle on all categories, even for basic tasks such as distinguishing microscopy modalities; ii) current specialist models fine-tuned on biomedical data often perform worse than generalist models; iii) fine-tuning in specific microscopy domains can cause catastrophic forgetting, eroding prior biomedical knowledge encoded in their base model. iv) weight interpolation between fine-tuned and pre-trained models offers one solution to forgetting and improves general performance across biomedical tasks. We release {\mu}-Bench under a permissive license to accelerate the research and development of microscopy foundation models.

Accurate classification of mode choice datasets is crucial for transportation planning and decision-making processes. However, conventional classification models often struggle to adequately capture the nuanced patterns of minority classes within these datasets, leading to sub-optimal accuracy. In response to this challenge, we present Ensemble Synthesizer (ENSY) which leverages probability distribution for data augmentation, a novel data model tailored specifically for enhancing classification accuracy in mode choice datasets. In our study, ENSY demonstrates remarkable efficacy by nearly quadrupling the F1 score of minority classes and improving overall classification accuracy by nearly 3%. To assess its performance comprehensively, we compare ENSY against various augmentation techniques including Random Oversampling, SMOTE-NC, and CTGAN. Through experimentation, ENSY consistently outperforms these methods across various scenarios, underscoring its robustness and effectiveness