Data Science

UMKC Announces New Master of Science in Artificial Intelligence

UMKC announces a new Master of Science in Artificial Intelligence program aimed at addressing workforce demand for AI expertise, set to l...

Scientists uncover new method to generate protein datasets for training AI

Accelerating science with AI and simulations

MIT Professor Rafael Gómez-Bombarelli discusses the transformative potential of AI in scientific research, emphasizing its role in materi...

[2501.15889] Adaptive Width Neural Networks

The paper introduces Adaptive Width Neural Networks, a novel approach that optimizes the width of neural network layers during training, ...

[2501.05633] Regularized Top-$k$: A Bayesian Framework for Gradient Sparsification

The paper presents a Bayesian framework for gradient sparsification called Regularized Top-k (RegTop-k), which improves convergence in di...

[2412.11439] Bayesian Flow Is All You Need to Sample Out-of-Distribution Chemical Spaces

The paper presents a Bayesian flow network, specifically the ChemBFN model, which effectively generates out-of-distribution chemical samp...

[2602.14881] Numerical exploration of the range of shape functionals using neural networks

This article presents a novel numerical framework for exploring shape functionals using neural networks, focusing on Blaschke–Santaló dia...

[2602.14879] CT-Bench: A Benchmark for Multimodal Lesion Understanding in Computed Tomography

CT-Bench introduces a benchmark dataset for multimodal lesion understanding in CT scans, featuring 20,335 lesions and a visual question a...

[2411.16085] Cautious Optimizers: Improving Training with One Line of Code

This article presents a new approach to optimizing training in machine learning by introducing a simple one-line modification to existing...

[2410.19412] VCDF: A Validated Consensus-Driven Framework for Time Series Causal Discovery

The paper presents VCDF, a consensus-driven framework for enhancing the robustness of time series causal discovery, improving stability a...

[2410.18784] Denoising diffusion probabilistic models are optimally adaptive to unknown low dimensionality

This paper explores the efficiency of denoising diffusion probabilistic models (DDPM) in adapting to unknown low dimensionality, proving ...

[2602.14834] Debiasing Central Fixation Confounds Reveals a Peripheral "Sweet Spot" for Human-like Scanpaths in Hard-Attention Vision

This paper explores the impact of central fixation bias on evaluating human-like scanpaths in vision models, proposing a new metric to im...

[2410.02081] MixLinear: Extreme Low Resource Multivariate Time Series Forecasting with 0.1K Parameters

MixLinear introduces an ultra-lightweight model for multivariate time series forecasting, achieving high accuracy with only 0.1K paramete...

[2602.14783] What hackers talk about when they talk about AI: Early-stage diffusion of a cybercrime innovation

This article explores how cybercriminals are discussing and utilizing artificial intelligence (AI) to enhance their operations, revealing...

[2408.11438] Benchmarking AI-based data assimilation to advance data-driven global weather forecasting

This article presents DABench, a benchmark for evaluating AI-based data assimilation methods in global weather forecasting, demonstrating...

[2407.11907] GraphFM: A generalist graph transformer that learns transferable representations across diverse domains

GraphFM introduces a scalable graph transformer that learns transferable representations across diverse domains, enhancing generalization...

[2406.12844] Synergizing Foundation Models and Federated Learning: A Survey

This survey explores the integration of Foundation Models (FMs) and Federated Learning (FL), termed Federated Foundation Models (FedFM), ...

[2405.14273] Exact Solution to Data-Driven Inverse Optimization of MILPs in Finite Time via Gradient-Based Methods

This article presents an exact solution to data-driven inverse optimization of mixed integer linear programs (MILPs) using gradient-based...

[2404.13895] Optimal Design for Human Preference Elicitation

The paper discusses optimal design strategies for eliciting human preferences, focusing on efficient methods for gathering high-quality f...

[2402.02644] Permutation-based Inference for Variational Learning of Directed Acyclic Graphs

This paper presents PIVID, a novel method for inferring distributions over permutations and directed acyclic graphs (DAGs) using variatio...

[2112.06251] Learning with Subset Stacking

The paper introduces a novel regression algorithm called Learning with Subset Stacking (LESS), which effectively learns from heterogeneou...

[2602.15021] Generalization from Low- to Moderate-Resolution Spectra with Neural Networks for Stellar Parameter Estimation: A Case Study with DESI

This article explores the use of neural networks for stellar parameter estimation, focusing on the transfer of data from low- to moderate...

[2602.15012] Cold-Start Personalization via Training-Free Priors from Structured World Models

This paper presents Pep, a novel approach for cold-start personalization that utilizes structured world models to improve user preference...