Machine Learning

MEow, a novel MaxEnt RL framework, achieves superior performance by unifying policy evaluation and improvement steps, enabling exact soft value function calculation without Monte Carlo approximation.

Adaptive MCMC with CNFs accelerates probabilistic inference by combining local and flow-informed transition kernels, achieving state-of-the-art results efficiently.

CoWorld: a novel model-based RL approach tackles offline visual RL challenges by using online simulators as testbeds, enabling flexible value estimation & mitigating overestimation bias for effective …

Boost continual learning with C-Flat: a novel, one-line-code optimizer creating flatter loss landscapes for enhanced stability and generalization across various continual learning scenarios.

Maia-2: A unified model for human-AI alignment in chess, coherently captures human play across skill levels, significantly improving AI-human alignment and paving the way for AI-guided teaching.

MADIFF: Offline multi-agent learning uses attention-based diffusion models to achieve effective coordination and teammate modeling, outperforming existing methods.

This paper establishes tight lower bounds for the uniform stability of gradient-based bilevel programming algorithms used for hyperparameter optimization, resolving a key open problem regarding the ti…

First optimal algorithms matching lower bounds for non-smooth convex decentralized optimization over time-varying networks are presented, substantially improving theoretical performance.

FRLC: a novel algorithm for low-rank optimal transport using latent coupling, enabling faster computation and better interpretability for diverse applications.

Low-precision local training, surprisingly, is sufficient for accurate federated learning, significantly reducing communication and computation costs.

Lorentz Geometric Algebra Transformer (L-GATr): A novel, scalable architecture for high-energy physics, achieving high-precision, data-efficient learning and outperforming existing methods on regressi…

LMSPS: a novel framework efficiently leverages long-range dependencies in large heterogeneous graphs by dynamically identifying effective meta-paths, mitigating computational costs and over-smoothing.

SSL vision encoders, while trained on massive datasets, surprisingly memorize individual data points. This paper introduces novel methods to precisely pinpoint this memorization within encoders at bot…

Local Superior Soups (LSS) significantly accelerates federated learning by efficiently merging pre-trained models, drastically cutting communication rounds without sacrificing accuracy.

CINDERELLA: a new algorithm achieves state-of-the-art no-regret bounds for continuous RL problems by exploiting local linearity.

LCSS, a novel score-matching method, enables efficient and high-quality image generation in score-based diffusion models by using Stein’s identity to bypass the computationally expensive Jacobian trac…

Greedy algorithms for linear contextual bandits achieve poly-logarithmic regret under the novel Local Anti-Concentration condition, expanding applicable distributions beyond Gaussians and uniforms.

LocalOMD: Adaptive OMD in extensive-form games achieves near-optimal sample complexity by using fixed sampling and local updates, reducing variance and generalizing well.

LM-HT SNN: A learnable multi-hierarchical threshold model dramatically improves SNN performance, achieving near-ANN accuracy through dynamic current regulation and seamless ANN-SNN conversion.

LLM-AutoDA: Automating data augmentation for long-tailed learning using large language models, significantly boosting model performance.