SPHERE applies a Parseval penalty to MoE policies in continual RL to maintain spectral plasticity, yielding 133% and 50% higher average success on MetaWorld and HumanoidBench versus unregularized MoE baselines.
Spectral normalization for generative adversarial networks
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
fields
cs.LG 2years
2026 2verdicts
UNVERDICTED 2representative citing papers
Synergistic active learning and input denoising reduces combined error in neural operators on viscous Burgers' equation from 15.42% to 2.04%, an 87% improvement over standard training.
citing papers explorer
-
SPHERE: Mitigating the Loss of Spectral Plasticity in Mixture-of-Experts for Deep Reinforcement Learning
SPHERE applies a Parseval penalty to MoE policies in continual RL to maintain spectral plasticity, yielding 133% and 50% higher average success on MetaWorld and HumanoidBench versus unregularized MoE baselines.
-
Beyond Uniform Sampling: Synergistic Active Learning and Input Denoising for Robust Neural Operators
Synergistic active learning and input denoising reduces combined error in neural operators on viscous Burgers' equation from 15.42% to 2.04%, an 87% improvement over standard training.