Search Results for author:
Found 51 papers, 23 papers with code
Preference Optimization as Probabilistic Inference
Existing preference optimization methods are mainly designed for directly learning from human feedback with the assumption that paired examples (preferred vs. dis-preferred) are available.
Game On: Towards Language Models as RL Experimenters
Additional examination of the ability of the system to build a growing library of skills, and to judge the progress of the training of those skills, also shows promising results, suggesting that the proposed architecture provides a potential recipe for fully automated mastery of tasks and domains for embodied agents.
Imitating Language via Scalable Inverse Reinforcement Learning
We focus on investigating the inverse reinforcement learning (IRL) perspective to imitation, extracting rewards and directly optimizing sequences instead of individual token likelihoods and evaluate its benefits for fine-tuning large language models.
Offline Actor-Critic Reinforcement Learning Scales to Large Models
We show that offline actor-critic reinforcement learning can scale to large models - such as transformers - and follows similar scaling laws as supervised learning.
GATS: Gather-Attend-Scatter
As the AI community increasingly adopts large-scale models, it is crucial to develop general and flexible tools to integrate them.
RoboCat: A Self-Improving Generalist Agent for Robotic Manipulation
With RoboCat, we demonstrate the ability to generalise to new tasks and robots, both zero-shot as well as through adaptation using only 100-1000 examples for the target task.
A Generalist Dynamics Model for Control
We investigate the use of transformer sequence models as dynamics models (TDMs) for control.
Leveraging Jumpy Models for Planning and Fast Learning in Robotic Domains
We conduct a set of experiments in the RGB-stacking environment, showing that planning with the learned skills and the associated model can enable zero-shot generalization to new tasks, and can further speed up training of policies via reinforcement learning.
A Generalist Agent
Inspired by progress in large-scale language modeling, we apply a similar approach towards building a single generalist agent beyond the realm of text outputs.
Ranked #1 on
Skill Generalization
on RGB-Stacking
How to Spend Your Robot Time: Bridging Kickstarting and Offline Reinforcement Learning for Vision-based Robotic Manipulation
Our analysis, both in simulation and in the real world, shows that our approach is the best across data budgets, while standard offline RL from teacher rollouts is surprisingly effective when enough data is given.
Revisiting Gaussian mixture critics in off-policy reinforcement learning: a sample-based approach
Actor-critic algorithms that make use of distributional policy evaluation have frequently been shown to outperform their non-distributional counterparts on many challenging control tasks.
Beyond Pick-and-Place: Tackling Robotic Stacking of Diverse Shapes
We study the problem of robotic stacking with objects of complex geometry.
Ranked #2 on
Skill Generalization
on RGB-Stacking
Evaluating model-based planning and planner amortization for continuous control
There is a widespread intuition that model-based control methods should be able to surpass the data efficiency of model-free approaches.
Collect & Infer -- a fresh look at data-efficient Reinforcement Learning
This position paper proposes a fresh look at Reinforcement Learning (RL) from the perspective of data-efficiency.
On Multi-objective Policy Optimization as a Tool for Reinforcement Learning: Case Studies in Offline RL and Finetuning
Many advances that have improved the robustness and efficiency of deep reinforcement learning (RL) algorithms can, in one way or another, be understood as introducing additional objectives or constraints in the policy optimization step.
Decoupled Exploration and Exploitation Policies for Sample-Efficient Reinforcement Learning
This causes BBE to be actively detrimental to policy learning in many control tasks.
Training Generative Adversarial Networks by Solving Ordinary Differential Equations
From this perspective, we hypothesise that instabilities in training GANs arise from the integration error in discretising the continuous dynamics.
Learning Dexterous Manipulation from Suboptimal Experts
Although in many cases the learning process could be guided by demonstrations or other suboptimal experts, current RL algorithms for continuous action spaces often fail to effectively utilize combinations of highly off-policy expert data and on-policy exploration data.
Local Search for Policy Iteration in Continuous Control
We demonstrate that additional computation spent on model-based policy improvement during learning can improve data efficiency, and confirm that model-based policy improvement during action selection can also be beneficial.
Critic Regularized Regression
Offline reinforcement learning (RL), also known as batch RL, offers the prospect of policy optimization from large pre-recorded datasets without online environment interaction.
Simple Sensor Intentions for Exploration
In particular, we show that a real robotic arm can learn to grasp and lift and solve a Ball-in-a-Cup task from scratch, when only raw sensor streams are used for both controller input and in the auxiliary reward definition.
Keep Doing What Worked: Behavior Modelling Priors for Offline Reinforcement Learning
In practice, however, standard off-policy algorithms fail in the batch setting for continuous control.
Keep Doing What Worked: Behavioral Modelling Priors for Offline Reinforcement Learning
In practice, however, standard off-policy algorithms fail in the batch setting for continuous control.
Continuous-Discrete Reinforcement Learning for Hybrid Control in Robotics
In contrast, we propose to treat hybrid problems in their 'native' form by solving them with hybrid reinforcement learning, which optimizes for discrete and continuous actions simultaneously.
Quinoa: a Q-function You Infer Normalized Over Actions
We present an algorithm for learning an approximate action-value soft Q-function in the relative entropy regularised reinforcement learning setting, for which an optimal improved policy can be recovered in closed form.
Imagined Value Gradients: Model-Based Policy Optimization with Transferable Latent Dynamics Models
Humans are masters at quickly learning many complex tasks, relying on an approximate understanding of the dynamics of their environments.
Model-based Reinforcement Learning
Reinforcement Learning
+3
V-MPO: On-Policy Maximum a Posteriori Policy Optimization for Discrete and Continuous Control
Some of the most successful applications of deep reinforcement learning to challenging domains in discrete and continuous control have used policy gradient methods in the on-policy setting.
Compositional Transfer in Hierarchical Reinforcement Learning
The successful application of general reinforcement learning algorithms to real-world robotics applications is often limited by their high data requirements.
General Reinforcement Learning
Hierarchical Reinforcement Learning
+5
Robust Reinforcement Learning for Continuous Control with Model Misspecification
We provide a framework for incorporating robustness -- to perturbations in the transition dynamics which we refer to as model misspecification -- into continuous control Reinforcement Learning (RL) algorithms.
Towards Automatically-Tuned Deep Neural Networks
Recent advances in AutoML have led to automated tools that can compete with machine learning experts on supervised learning tasks.
Self-supervised Learning of Image Embedding for Continuous Control
Operating directly from raw high dimensional sensory inputs like images is still a challenge for robotic control.
Relative Entropy Regularized Policy Iteration
Our algorithm draws on connections to existing literature on black-box optimization and 'RL as an inference' and it can be seen either as an extension of the Maximum a Posteriori Policy Optimisation algorithm (MPO) [Abdolmaleki et al., 2018a], or as an extension of Trust Region Covariance Matrix Adaptation Evolutionary Strategy (CMA-ES) [Abdolmaleki et al., 2017b; Hansen et al., 1997] to a policy iteration scheme.
Maximum a Posteriori Policy Optimisation
We introduce a new algorithm for reinforcement learning called Maximum aposteriori Policy Optimisation (MPO) based on coordinate ascent on a relative entropy objective.
Graph networks as learnable physics engines for inference and control
Understanding and interacting with everyday physical scenes requires rich knowledge about the structure of the world, represented either implicitly in a value or policy function, or explicitly in a transition model.
Ranked #9 on
Weather Forecasting
on SD
Learning by Playing - Solving Sparse Reward Tasks from Scratch
We propose Scheduled Auxiliary Control (SAC-X), a new learning paradigm in the context of Reinforcement Learning (RL).
Learning an Embedding Space for Transferable Robot Skills
We present a method for reinforcement learning of closely related skills that are parameterized via a skill embedding space.
Deep learning with convolutional neural networks for EEG decoding and visualization
PLEASE READ AND CITE THE REVISED VERSION at Human Brain Mapping: http://onlinelibrary. wiley. com/doi/10. 1002/hbm. 23730/full Code available here: https://github. com/robintibor/braindecode
Deep Reinforcement Learning with Successor Features for Navigation across Similar Environments
We propose a successor feature based deep reinforcement learning algorithm that can learn to transfer knowledge from previously mastered navigation tasks to new problem instances.
Asynchronous Stochastic Gradient MCMC with Elastic Coupling
We consider parallel asynchronous Markov Chain Monte Carlo (MCMC) sampling for problems where we can leverage (stochastic) gradients to define continuous dynamics which explore the target distribution.
Bayesian Optimization with Robust Bayesian Neural Networks
Bayesian optimization is a prominent method for optimizing expensive to evaluate black-box functions that is prominently applied to tuning the hyperparameters of machine learning algorithms.
Unsupervised and Semi-supervised Learning with Categorical Generative Adversarial Networks
Our approach is based on an objective function that trades-off mutual information between observed examples and their predicted categorical class distribution, against robustness of the classifier to an adversarial generative model.
Ranked #8 on
Unsupervised MNIST
on MNIST
Multimodal Deep Learning for Robust RGB-D Object Recognition
Robust object recognition is a crucial ingredient of many, if not all, real-world robotics applications.
Embed to Control: A Locally Linear Latent Dynamics Model for Control from Raw Images
We introduce Embed to Control (E2C), a method for model learning and control of non-linear dynamical systems from raw pixel images.
Learning to Generate Chairs With Convolutional Neural Networks
We train a generative convolutional neural network which is able to generate images of objects given object type, viewpoint, and color.
Supplementary Material for Efficient and Robust Automated Machine Learning
Supplementary Material for Efficient and Robust Automated Machine Learning
Striving for Simplicity: The All Convolutional Net
Most modern convolutional neural networks (CNNs) used for object recognition are built using the same principles: Alternating convolution and max-pooling layers followed by a small number of fully connected layers.
Ranked #123 on
Image Classification
on CIFAR-10
Discriminative Unsupervised Feature Learning with Convolutional Neural Networks
Current methods for training convolutional neural networks depend on large amounts of labeled samples for supervised training.
Ranked #83 on
Image Classification
on STL-10
Learning to Generate Chairs, Tables and Cars with Convolutional Networks
We train generative 'up-convolutional' neural networks which are able to generate images of objects given object style, viewpoint, and color.
Discriminative Unsupervised Feature Learning with Exemplar Convolutional Neural Networks
While such generic features cannot compete with class specific features from supervised training on a classification task, we show that they are advantageous on geometric matching problems, where they also outperform the SIFT descriptor.
Improving Deep Neural Networks with Probabilistic Maxout Units
We present a probabilistic variant of the recently introduced maxout unit.
Ranked #184 on
Image Classification
on CIFAR-100
Unsupervised feature learning by augmenting single images
We then extend these trivial one-element classes by applying a variety of transformations to the initial 'seed' patches.
