Machine Learning

LOLIPOP: A novel algorithm achieving near-optimal regret for offline contextual Markov Decision Processes (CMDPs) using only O(H log T) offline density estimation oracle calls.

This paper introduces Offline Behavior Distillation (OBD) to synthesize compact expert behavioral data from massive sub-optimal RL data, enabling faster policy learning.

DARAIL, a novel algorithm, tackles off-dynamics reinforcement learning by combining reward modification with imitation learning to transfer a learned policy from a source to a target domain. This app…

Model-free policy gradient methods using occupancy functions are developed for online and offline RL, achieving computational efficiency and handling arbitrary data distributions.

OASIS, a novel data-centric approach, shapes offline data distributions toward safer, higher-reward policies using a conditional diffusion model, outperforming existing offline safe RL methods.

This paper presents a novel, efficient method for Jacobian nuclear norm regularization in deep learning, replacing computationally expensive SVDs with equivalent Frobenius norm computations, thereby e…

Normalize-and-Project (NaP) boosts reinforcement learning by stabilizing layer normalization, preventing plasticity loss, and enabling effective learning rate control.

Nonstationary Sparse Spectral Permanental Process (NSSPP) enhances point process modeling by using sparse spectral representations, enabling flexible, efficient, nonstationary kernel learning.

Adaptive algorithm selection for noisy ICA is achieved via a novel nonparametric independence score, improving accuracy and efficiency.

Overparameterized ConvResNets surprisingly excel at prediction; this study proves they efficiently learn smooth functions on low-dimensional manifolds, avoiding the curse of dimensionality.

This paper proposes a novel federated learning algorithm for nonconvex problems on compact smooth manifolds, achieving both computational and communication efficiency while mitigating client drift.

AI models struggle with changing data; this paper introduces Soft Resets, a novel learning approach that uses an adaptive drift to gracefully guide parameters toward initialization, improving adaptabi…

New non-parametric classifiers using expand-and-sparsify representation achieve minimax-optimal convergence, adapting to low-dimensional manifold structure.

Non-Euclidean Mixture Model (NMM-GNN) outperforms existing methods by using spherical and hyperbolic spaces to model homophily and social influence in social network embedding, improving link predicti…

This paper rigorously analyzes biased adaptive stochastic gradient descent (SGD), proving convergence to critical points for non-convex functions even with biased gradient estimations. The analysis c…

Noether’s Razor learns conserved quantities and symmetries directly from data via Bayesian model selection, improving dynamical systems modeling accuracy and generalizability.

A novel stochastic bandit algorithm using soft tree ensemble models achieves lower cumulative regret than existing ReLU-based neural bandit algorithms, offering a constrained yet effective hypothesis …

Deep RL agents trained under non-stationarity suffer performance collapse due to representation degradation; this work reveals this in PPO and introduces Proximal Feature Optimization (PFO) to mitigat…

AI agents learn better with well-designed training environments. This paper reveals flaws in current environment-selection methods and introduces Sampling for Learnability (SFL), a new approach that …

Newton Losses enhance training of neural networks with complex objectives by using second-order information from loss functions, achieving significant performance gains.