Data Science

Harvard opens more free online courses in AI, data science, programming: Check full list and direct links

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...

[D] Offering licensed Indian language speech datasets (with explicit contributor consent)

Hi everyone, I run a small data initiative where we collect speech datasets in multiple Indian languages directly from contributors who p...

[2602.18718] Stochastic Gradient Variational Inference with Price's Gradient Estimator from Bures-Wasserstein to Parameter Space

This paper presents advancements in Stochastic Gradient Variational Inference (SGVI) using Price's Gradient Estimator, demonstrating comp...

[2602.19177] Next Reply Prediction X Dataset: Linguistic Discrepancies in Naively Generated Content

The paper introduces the Next Reply Prediction X Dataset, addressing linguistic discrepancies in content generated by Large Language Mode...

[2602.18715] A Data-Driven Method to Map the Functional Organisation of Human Brain White Matter

This article presents a data-driven method to map the functional organization of human brain white matter, integrating diffusion and func...

[2602.19171] HistCAD: Geometrically Constrained Parametric History-based CAD Dataset

The paper presents HistCAD, a comprehensive dataset for parametric CAD modeling that incorporates geometric constraints and functional se...

[2602.19153] Constrained Diffusion for Accelerated Structure Relaxation of Inorganic Solids with Point Defects

This article presents a novel generative framework for simulating point defects in inorganic solids, enhancing structure relaxation proce...

[2602.19156] Artefact-Aware Fungal Detection in Dermatophytosis: A Real-Time Transformer-Based Approach for KOH Microscopy

This study presents a transformer-based framework for detecting fungal elements in dermatophytosis using KOH microscopy, achieving high a...

[2602.18642] Auto Quantum Machine Learning for Multisource Classification

The paper presents an automated quantum machine learning (AQML) approach for multisource classification, demonstrating improved accuracy ...

[2602.19138] CRCC: Contrast-Based Robust Cross-Subject and Cross-Site Representation Learning for EEG

The paper presents CRCC, a novel framework for improving EEG-based neural decoding models' generalization across different acquisition si...

[2602.18573] Multiclass Calibration Assessment and Recalibration of Probability Predictions via the Linear Log Odds Calibration Function

The paper presents a novel method for assessing and recalibrating probability predictions in multiclass classification tasks, addressing ...

[2602.18525] Do Generative Metrics Predict YOLO Performance? An Evaluation Across Models, Augmentation Ratios, and Dataset Complexity

This paper evaluates the effectiveness of generative metrics in predicting the performance of YOLO object detection models across various...

[2602.19087] Detecting Cybersecurity Threats by Integrating Explainable AI with SHAP Interpretability and Strategic Data Sampling

This article presents a novel framework for detecting cybersecurity threats by integrating Explainable AI (XAI) with SHAP interpretabilit...

[2602.18487] The Million-Label NER: Breaking Scale Barriers with GLiNER bi-encoder

The paper presents GLiNER-bi-Encoder, a new architecture for Named Entity Recognition (NER) that enhances efficiency and scalability, ena...

[2602.18489] DCInject: Persistent Backdoor Attacks via Frequency Manipulation in Personal Federated Learning

The paper presents DCInject, a novel backdoor attack method targeting personalized federated learning (PFL) systems, demonstrating high a...

[2602.18482] Boltzmann Generators for Condensed Matter via Riemannian Flow Matching

This article presents a novel approach using Riemannian flow matching to enhance Boltzmann generators for sampling equilibrium distributi...

[2602.19028] The Metaphysics We Train: A Heideggerian Reading of Machine Learning

This paper explores machine learning through a Heideggerian lens, highlighting insights on algorithmic opacity, the limitations of calcul...

[2602.19025] Routing-Aware Explanations for Mixture of Experts Graph Models in Malware Detection

This article presents a novel approach to malware detection using Mixture-of-Experts (MoE) graph models, emphasizing routing-aware explan...

[2602.19022] An interpretable framework using foundation models for fish sex identification

The paper presents FishProtoNet, a non-invasive computer vision framework for accurately identifying the sex of delta smelt, an endangere...

[2602.18962] NeuroWise: A Multi-Agent LLM "Glass-Box" System for Practicing Double-Empathy Communication with Autistic Partners

NeuroWise is a multi-agent LLM system designed to enhance double-empathy communication between neurotypical and autistic individuals, dem...

[2602.20152] Behavior Learning (BL): Learning Hierarchical Optimization Structures from Data

The paper introduces Behavior Learning (BL), a machine learning framework that learns interpretable optimization structures from data, en...

[2602.20111] Reliable Abstention under Adversarial Injections: Tight Lower Bounds and New Upper Bounds

This paper explores reliable abstention in online learning under adversarial injections, presenting new lower and upper bounds for error ...