23 citations · 75 across the 24 of their papers we have counts for
20 papers · 1 filter
Value-Distributional Model-Based Reinforcement Learning
Carlos E. Luis, Alessandro G. Bottero, Julia Vinogradska +2
Quantifying uncertainty about a policy's long-term performance is important to solve sequential decision-making tasks. We study the problem from a model-based Bayesian reinforcemen…
Function-Space Regularization for Deep Bayesian Classification
Jihao Andreas Lin, Joe Watson, Pascal Klink +1
Bayesian deep learning approaches assume model parameters to be latent random variables and infer posterior distributions to quantify uncertainty, increase safety and trust, and pr…
Cheap and Deterministic Inference for Deep State-Space Models of Interacting Dynamical Systems
Andreas Look, Melih Kandemir, Barbara Rakitsch +1
Graph neural networks are often used to model interacting dynamical systems since they gracefully scale to systems with a varying and high number of agents. While there has been mu…
Diminishing Return of Value Expansion Methods in Model-Based Reinforcement Learning
Daniel Palenicek, Michael Lutter, Joao Carvalho +1
Model-based reinforcement learning is one approach to increase sample efficiency. However, the accuracy of the dynamics model and the resulting compounding error over modelled traj…
Model-Based Uncertainty in Value Functions
Carlos E. Luis, Alessandro G. Bottero, Julia Vinogradska +2
We consider the problem of quantifying uncertainty over expected cumulative rewards in model-based reinforcement learning. In particular, we focus on characterizing the variance ov…
LS-IQ: Implicit Reward Regularization for Inverse Reinforcement Learning
Firas Al-Hafez, Davide Tateo, Oleg Arenz +2
Recent methods for imitation learning directly learn a -function using an implicit reward formulation rather than an explicit reward function. However, these methods generally r…
Information-Theoretic Safe Exploration with Gaussian Processes
Alessandro G. Bottero, Carlos E. Luis, Julia Vinogradska +2
We consider a sequential decision making task where we are not allowed to evaluate parameters that violate an a priori unknown (safety) constraint. A common approach is to place a…
PAC-Bayes Bounds for Bandit Problems: A Survey and Experimental Comparison
Hamish Flynn, David Reeb, Melih Kandemir +1
PAC-Bayes has recently re-emerged as an effective theory with which one can derive principled learning algorithms with tight performance guarantees. However, applications of PAC-Ba…
Variational Hierarchical Mixtures for Probabilistic Learning of Inverse Dynamics
Hany Abdulsamad, Peter Nickl, Pascal Klink +1
Well-calibrated probabilistic regression models are a crucial learning component in robotics applications as datasets grow rapidly and tasks become more complex. Unfortunately, cla…
Structured Q-learning For Antibody Design
Alexander I. Cowen-Rivers, Philip John Gorinski, Aivar Sootla +5
Optimizing combinatorial structures is core to many real-world problems, such as those encountered in life sciences. For example, one of the crucial steps involved in antibody desi…
Rollout Total Correlation for Deep Reinforcement Learning
Bang You, Huaping Liu, Jingming Xie +3
Learning task-relevant representations is crucial for reinforcement learning. Recent approaches aim to learn such representations by improving the temporal consistency in the obser…
Revisiting Model-based Value Expansion
Daniel Palenicek, Michael Lutter, Jan Peters
Model-based value expansion methods promise to improve the quality of value function targets and, thereby, the effectiveness of value function learning. However, to date, these met…
Dimensionality Reduction and Prioritized Exploration for Policy Search
Marius Memmel, Puze Liu, Davide Tateo +1
Black-box policy optimization is a class of reinforcement learning algorithms that explores and updates the policies at the parameter level. This class of algorithms is widely appl…
An Analysis of Measure-Valued Derivatives for Policy Gradients
Joao Carvalho, Jan Peters
Reinforcement learning methods for robotics are increasingly successful due to the constant development of better policy gradient techniques. A precise (low variance) and accurate…
PAC-Bayesian Lifelong Learning For Multi-Armed Bandits
Hamish Flynn, David Reeb, Melih Kandemir +1
We present a PAC-Bayesian analysis of lifelong learning. In the lifelong learning problem, a sequence of learning tasks is observed one-at-a-time, and the goal is to transfer infor…
Distilled Domain Randomization
Julien Brosseit, Benedikt Hahner, Fabio Muratore +2
Deep reinforcement learning is an effective tool to learn robot control policies from scratch. However, these methods are notorious for the enormous amount of required training dat…
Efficient Stochastic Optimal Control through Approximate Bayesian Input Inference
Joe Watson, Hany Abdulsamad, Rolf Findeisen +1
Optimal control under uncertainty is a prevailing challenge for many reasons. One of the critical difficulties lies in producing tractable solutions for the underlying stochastic o…
HEBO Pushing The Limits of Sample-Efficient Hyperparameter Optimisation
Alexander I. Cowen-Rivers, Wenlong Lyu, Rasul Tutunov +8
In this work we rigorously analyse assumptions inherent to black-box optimisation hyper-parameter tuning tasks. Our results on the Bayesmark benchmark indicate that heteroscedastic…
A Deterministic Approximation to Neural SDEs
Andreas Look, Melih Kandemir, Barbara Rakitsch +1
Neural Stochastic Differential Equations (NSDEs) model the drift and diffusion functions of a stochastic process as neural networks. While NSDEs are known to make accurate predicti…
PAC-Bayesian Analysis of the Exploration-Exploitation Trade-off
Yevgeny Seldin, Nicolò Cesa-Bianchi, François Laviolette +3
We develop a coherent framework for integrative simultaneous analysis of the exploration-exploitation and model order selection trade-offs. We improve over our preceding results on…