GSBR raises robotic RL success from 33% to 95%
Geometric Support Behavior Regularization allows base-policy improvements while penalizing actions outside support on 19 robotic tasks.
Stories per day · 30 days
Relatedbase-policies1inverse-problems1medical-imaging1photoacoustic-tomography1reinforcement-learning1robotic-benchmarks1untrained-cnn1
11 stories, newest first · one card per story
Stories tagged #regularization
Untrained CNN and coefficient-space regularization for joint reconstruction in photoacoustic tomography
An untrained CNN with coefficient regularization jointly reconstructs photoacoustic pressure and sound speed.
BiasFlow: Geometric Monitoring and Backbone Regularization for Spurious Feature Reliance
BiasFlow regularizes backbone centroids, lifting worst-group accuracy by up to 26 points on UrbanCars.
Provable Benefits of Regularization: Fast Rates for Adversarial Imitation Learning
Theory paper proves fast finite-sample rates for dually regularized adversarial imitation learning.
Generalization behavior of OPTQ and the role of regularization
New bounds show OPTQ generalization depends on regularization lambda, and a revised lambda choice beats prior recommendations.
RRSI: Regularized Recursive Self-Improvement of Agent Harnesses
Google Research's RRSI regularizes recursive agent-harness edits, lifting held-out scores while cutting policy tokens.
RRSI: Regularized Recursive Self-Improvement of Agent Harnesses
Google Research's RRSI regularizes recursive agent-harness evolution, lifting out-of-distribution scores while cutting policy tokens 30%.
Double descent is the principle of least action
A statistical mechanics analysis explains double descent: finite-time diffusion induces effective weight decay that regularizes models as parameters grow.
A Unified and Constrained View of Regularization-Based Robust Reinforcement Learning
Paper unifies regularization-based robust RL methods via new performance-gap upper bounds and jointly learned Lagrange multipliers.
Data Scarcity and Model Sparsity: Mixtures-of-Experts Overfit More to Repeated Data
Study finds Mixture-of-Experts models overfit faster than dense Transformers under repeated training data, with degradation tied to total parameter sparsity.
On the Regularization Landscape for the Linear Recommendation Models
Study shows leading linear recommendation models reduce to nuclear-norm or Frobenius-norm regularization, with two new closed-form low-rank solutions proposed.