Education
Deep Learning Through the Lens of Example Difficulty
Existing work on understanding deep learning often employs measures that compress all data-dependent information into a few numbers.
Information is Power: Intrinsic Control via Information Capture
Figure 1: The agent uses a latent state space model to represent beliefs about the world, including dynamic objects like the goat. The blue window represents the agent's field-of-view, which defines the extent of the
Provably Feedback-Efficient Reinforcement Learning via Active Reward Learning
An appropriate reward function is of paramount importance in specifying a task in reinforcement learning (RL).
Learning from Distributed Users in Contextual Linear Bandits Without Sharing the Context
In this paper, we develop algorithms that support the deployment of contextual linear bandits in distributed settings.
Online Learning in MDPs with Linear Function Approximation and Bandit Feedback
Consequently, the state of the environment changes according to the transition function of the underlying MDP, as a function of the previous state and the action taken by the learner.
Memory Efficient Continual Learning with Transformers
To address the issue of incremental fine-tuning of pre-trained Transformers in the sequential learning setting without CF, we propose Adaptive Distillation of Adapters (ADA).