Search Results for author:
Found 47 papers, 18 papers with code
Beyond Uniform Sampling: Offline Reinforcement Learning with Imbalanced Datasets
We argue this is due to an assumption made by current offline RL algorithms of staying close to the trajectories in the dataset.
Consciousness-Inspired Spatio-Temporal Abstractions for Better Generalization in Reinforcement Learning
Inspired by human conscious planning, we propose Skipper, a model-based reinforcement learning framework utilizing spatio-temporal abstractions to generalize better in novel situations.
Harnessing Mixed Offline Reinforcement Learning Datasets via Trajectory Weighting
This re-weighted sampling strategy may be combined with any offline RL algorithm.
Think Before You Act: Decision Transformers with Internal Working Memory
We argue that this inefficiency stems from the forgetting phenomenon, in which a model memorizes its behaviors in parameters throughout training.
Behavior Prior Representation learning for Offline Reinforcement Learning
Our method, Behavior Prior Representation (BPR), learns state representations with an easy-to-integrate objective based on behavior cloning of the dataset: we first learn a state representation by mimicking actions from the dataset, and then train a policy on top of the fixed representation, using any off-the-shelf Offline RL algorithm.
Discrete Factorial Representations as an Abstraction for Goal Conditioned Reinforcement Learning
Goal-conditioned reinforcement learning (RL) is a promising direction for training agents that are capable of solving multiple tasks and reach a diverse set of objectives.
Contrastive Multimodal Learning for Emergence of Graphical Sensory-Motor Communication
In this paper, we investigate whether artificial agents can develop a shared language in an ecological setting where communication relies on a sensory-motor channel.
Incorporating Explicit Uncertainty Estimates into Deep Offline Reinforcement Learning
In this work, we develop a novel method for incorporating scalable uncertainty estimates into an offline reinforcement learning algorithm called deep-SPIBB that extends the SPIBB family of algorithms to environments with larger state and action spaces.
Using Representation Expressiveness and Learnability to Evaluate Self-Supervised Learning Methods
We address the problem of evaluating the quality of self-supervised learning (SSL) models without access to supervised labels, while being agnostic to the architecture, learning algorithm or data manipulation used during training.
When does return-conditioned supervised learning work for offline reinforcement learning?
Several recent works have proposed a class of algorithms for the offline reinforcement learning (RL) problem that we will refer to as return-conditioned supervised learning (RCSL).
Non-Markovian policies occupancy measures
A central object of study in Reinforcement Learning (RL) is the Markovian policy, in which an agent's actions are chosen from a memoryless probability distribution, conditioned only on its current state.
One-Shot Learning from a Demonstration with Hierarchical Latent Language
Humans have the capability, aided by the expressive compositionality of their language, to learn quickly by demonstration.
Beyond the Policy Gradient Theorem for Efficient Policy Updates in Actor-Critic Algorithms
To increase the unlearning speed, we study a novel policy update: the gradient of the cross-entropy loss with respect to the action maximizing $q$, but find that such updates may lead to a decrease in value.
On the Convergence of SARSA with Linear Function Approximation
SARSA, a classical on-policy control algorithm for reinforcement learning, is known to chatter when combined with linear function approximation: SARSA does not diverge but oscillates in a bounded region.
Dr Jekyll & Mr Hyde: the strange case of off-policy policy updates
To implement the principles prescribed by our theory, we propose an agent, Dr Jekyll & Mr Hyde (J&H), with a double personality: Dr Jekyll purely exploits while Mr Hyde purely explores.
Global Optimality and Finite Sample Analysis of Softmax Off-Policy Actor Critic under State Distribution Mismatch
In this paper, we establish the global optimality and convergence rate of an off-policy actor critic algorithm in the tabular setting without using density ratio to correct the discrepancy between the state distribution of the behavior policy and that of the target policy.
Batched Bandits with Crowd Externalities
In Batched Multi-Armed Bandits (BMAB), the policy is not allowed to be updated at each time step.
Learnability and Expressiveness in Self-Supervised Learning
In this work, we argue that representations induced by self-supervised learning (SSL) methods should both be expressive and learnable.
Dr Jekyll and Mr Hyde: the Strange Case of Off-Policy Policy Updates
To implement the principles prescribed by our theory, we propose an agent, Dr Jekyll & Mr Hyde (JH), with a double personality: Dr Jekyll purely exploits while Mr Hyde purely explores.
The Emergence of the Shape Bias Results from Communicative Efficiency
By the age of two, children tend to assume that new word categories are based on objects' shape, rather than their color or texture; this assumption is called the shape bias.
Multi-Objective SPIBB: Seldonian Offline Policy Improvement with Safety Constraints in Finite MDPs
We study the problem of Safe Policy Improvement (SPI) under constraints in the offline Reinforcement Learning (RL) setting.
A Deeper Look at Discounting Mismatch in Actor-Critic Algorithms
In the second scenario, we consider optimizing a discounted objective ($\gamma < 1$) and propose to interpret the omission of the discounting in the actor update from an auxiliary task perspective and provide supporting empirical results.
Reinforcement Learning Framework for Deep Brain Stimulation Study
Malfunctioning neurons in the brain sometimes operate synchronously, reportedly causing many neurological diseases, e. g. Parkinson's.
Learning Dynamic Belief Graphs to Generalize on Text-Based Games
Playing text-based games requires skills in processing natural language and sequential decision making.
Building Dynamic Knowledge Graphs from Text-based Games
We are interested in learning how to update Knowledge Graphs (KG) from text.
Safe Policy Improvement with an Estimated Baseline Policy
Previous work has shown the unreliability of existing algorithms in the batch Reinforcement Learning setting, and proposed the theoretically-grounded Safe Policy Improvement with Baseline Bootstrapping (SPIBB) fix: reproduce the baseline policy in the uncertain state-action pairs, in order to control the variance on the trained policy performance.
Safe Policy Improvement with Soft Baseline Bootstrapping
Batch Reinforcement Learning (Batch RL) consists in training a policy using trajectories collected with another policy, called the behavioural policy.
Budgeted Reinforcement Learning in Continuous State Space
A Budgeted Markov Decision Process (BMDP) is an extension of a Markov Decision Process to critical applications requiring safety constraints.
Decentralized Exploration in Multi-Armed Bandits -- Extended version
We consider the decentralized exploration problem: a set of players collaborate to identify the best arm by asynchronously interacting with the same stochastic environment.
Counting to Explore and Generalize in Text-based Games
We propose a recurrent RL agent with an episodic exploration mechanism that helps discovering good policies in text-based game environments.
Safe Policy Improvement with Baseline Bootstrapping
Finally, we implement a model-free version of SPIBB and show its benefits on a navigation task with deep RL implementation called SPIBB-DQN, which is, to the best of our knowledge, the first RL algorithm relying on a neural network representation able to train efficiently and reliably from batch data, without any interaction with the environment.
The Complex Negotiation Dialogue Game
This position paper formalises an abstract model for complex negotiation dialogue.
Hybrid Reward Architecture for Reinforcement Learning
One of the main challenges in reinforcement learning (RL) is generalisation.
Multi-Advisor Reinforcement Learning
We consider tackling a single-agent RL problem by distributing it to $n$ learners.
Reinforcement Learning Algorithm Selection
This paper formalises the problem of online algorithm selection in the context of Reinforcement Learning.
Separation of Concerns in Reinforcement Learning
In this paper, we propose a framework for solving a single-agent task by using multiple agents, each focusing on different aspects of the task.
NASTIA: Negotiating Appointment Setting Interface
Appointment scheduling is a hybrid task halfway between slot-filling and negotiation.
DINASTI: Dialogues with a Negotiating Appointment Setting Interface
NASTIA is a reinforcement learning-based system.
