Journals, conferences & preprints
Each entry opens a full write-up: the idea, the method, the figures, and a citation you can copy.
Recommended reading
We propose contraction-aware reinforcement learning (CARL), which simultaneously learns control contraction metrics (CCMs) and optimizes a policy against rewards defined by those metrics, yielding stable and statistically robust path tracking for high-dimensional nonlinear systems.
Everything
Filter by topic. The filter travels in the URL, so a filtered view can be linked.
TL;DR — We propose contraction-aware reinforcement learning (CARL), which simultaneously learns control contraction metrics (CCMs) and optimizes a policy against rewards defined by those metrics, yielding stable and statistically robust path tracking for high-dimensional nonlinear systems.
TL;DR — We propose to learn transition dynamics and reward function using causal normalizing flow model for out-of-distribution adaptation of a policy.
TL;DR — We propose a macro-action discovery method and use it in a hierarchical algorithm for solving complex meta-RL problems.
TL;DR — We propose to use K-Mean clustering algorithm to measure the sparsity of each data point and use those measures to overestimate the probable safety violation for safe deployment of a policy.
TL;DR — We propose a policy optimization framework that provides adaptable safety guarantees on unseen tasks by viewing constrained reinforcement learning through the lens of meta-learning.
TL;DR — We propose Intrinsic Reward Policy Optimization (IRPO), a novel framework leveraging a surrogate policy gradient to overcome credit assignment and sample inefficiency in sparse-reward environments.
Elsewhere
Citation counts and the canonical record live off-site.