5 papers
FIRE: Frobenius-Isometry Reinitialization for Balancing the Stability-Plasticity Tradeoff
Isaac Han, Sangyeon Park, Seungwon Oh +3
Deep neural networks trained on nonstationary data must balance stability (i.e., retaining prior knowledge) and plasticity (i.e., adapting to new tasks). Standard reinitialization…
Prism: Spectral Parameter Sharing for Multi-Agent Reinforcement Learning
Kyungbeom Kim, Seungwon Oh, Kyung-Joong Kim
Parameter sharing is a key strategy in multi-agent reinforcement learning (MARL) for improving scalability, yet conventional fully shared architectures often collapse into homogene…
A Humanoid Visual-Tactile-Action Dataset for Contact-Rich Manipulation
Eunju Kwon, Seungwon Oh, In-Chang Baek +5
Contact-rich manipulation has become increasingly important in robot learning. However, previous studies on robot learning datasets have focused on rigid objects and underrepresent…
Recovering Plasticity of Neural Networks via Soft Weight Rescaling
Seungwon Oh, Sangyeon Park, Isaac Han +1
Recent studies have shown that as training progresses, neural networks gradually lose their capacity to learn new information, a phenomenon known as plasticity loss. An unbounded w…
Activation by Interval-wise Dropout: A Simple Way to Prevent Neural Networks from Plasticity Loss
Sangyeon Park, Isaac Han, Seungwon Oh +1
Plasticity loss, a critical challenge in neural network training, limits a model's ability to adapt to new tasks or shifts in data distribution. This paper introduces AID (Activati…