#theoretical-ml News & Analysis

6 articles tagged with #theoretical-ml. AI-curated summaries with sentiment analysis and key takeaways from 50+ sources.

6 articles

AINeutralarXiv – CS AI · Jun 56/10

🧠

Inverse Entropic Optimal Transport Solves Semi-supervised Learning via Data Likelihood Maximization

Researchers propose EBiEOT, a novel semi-supervised learning framework that leverages both paired and unpaired data through likelihood maximization and inverse entropic optimal transport. The method demonstrates universal approximation properties and provides an end-to-end algorithm for learning conditional distributions, with potential applications in domain translation and other data-scarce scenarios.

AINeutralarXiv – CS AI · Jun 26/10

🧠

Probabilistic Performance Guarantees for Multi-Task Reinforcement Learning

Researchers present a new theoretical framework for multi-task reinforcement learning that computes high-confidence performance guarantees on unseen tasks by combining per-task confidence bounds with task-level generalization. The approach addresses a critical gap in deploying RL policies in safety-critical applications where formal performance assurances are essential.

AIBullisharXiv – CS AI · May 126/10

🧠

A Unified Pair-GRPO Family: From Implicit to Explicit Preference Constraints for Stable and General RL Alignment

Researchers propose Pair-GRPO, a unified theoretical framework for LLM alignment that addresses instability and interpretability issues in reinforcement learning from human preferences. The method introduces Soft-Pair-GRPO and Hard-Pair-GRPO variants with proven gradient equivalence, monotonic policy improvement, and superior performance on standard benchmarks.

AINeutralarXiv – CS AI · May 116/10

🧠

$f$-Divergence Regularized RLHF: Two Tales of Sampling and Unified Analyses

Researchers present a unified theoretical framework for f-divergence regularized Reinforcement Learning from Human Feedback (RLHF), moving beyond the standard reverse KL approach. The work introduces two novel algorithms with provable efficiency guarantees, achieving O(log T) regret bounds and establishing the first theoretical performance guarantees for online RLHF under general f-divergence regularization.

AINeutralarXiv – CS AI · May 116/10

🧠

Flat Channels to Infinity in Neural Loss Landscapes

Researchers identify and characterize 'channels to infinity' in neural network loss landscapes—flat regions where neurons diverge to extreme values while converging to shared weight vectors. These structures, which gradient-based optimizers frequently reach, functionally collapse to gated linear units and reveal surprising computational properties of fully connected layers.

AINeutralarXiv – CS AI · May 96/10

🧠

Consistent Geometric Deep Learning via Hilbert Bundles and Cellular Sheaves

Researchers introduce HilbNets, a novel deep learning framework that handles infinite-dimensional signals (like time series and probability distributions) on irregular domains using Hilbert bundles and cellular sheaves. The work provides theoretical convergence guarantees and demonstrates that discretized networks maintain consistency across different data sampling schemes, advancing geometric deep learning theory.