Bayesian Learning
Distributional Successor Features Enable Zero-Shot Policy Optimization
Intelligent agents must be generalists, capable of quickly adapting to various tasks.
Imitation learning (IL) aims to mimic the behavior of an expert in a sequential decision making task by learning from demonstrations, and has been widely applied
Full experimental details are provided in Appendix C.
Principled Probabilistic Imaging using Diffusion Models as Plug-and-Play Priors
The black hole imaging problem is non-convex and highly ill-posed due to severe noise corruption and measurement sparsity.
Conditional Density Estimation with Histogram Trees
This makes CDE particularly useful in critical application domains. However, interpretable CDE methods are understudied.
Label Noise: Ignorance Is Bliss
We establish a new theoretical framework for learning under multi-class, instance-dependent label noise. This framework casts learning with label noise as a form of domain adaptation, in particular, domain adaptation under posterior drift.
Model and Feature Diversity for Bayesian Neural Networks in Mutual Learning
Utilizing mutual learning can effectively enhance the performance of peer BNNs.
Meta-Learning Universal Priors Using Non-Injective Change of Variables
Critical in this context is the prior knowledge accumulated from related tasks.