Vibhavari Dasagi
YOU?
Author Swipe
View article: Steering Language Models with Game-Theoretic Solvers
Steering Language Models with Game-Theoretic Solvers Open
Mathematical models of interactions among rational agents have long been studied in game theory. However these interactions are often over a small set of discrete game actions which is very different from how humans communicate in natural …
View article: Bayesian controller fusion: Leveraging control priors in deep reinforcement learning for robotics
Bayesian controller fusion: Leveraging control priors in deep reinforcement learning for robotics Open
We present Bayesian Controller Fusion (BCF): a hybrid control strategy that combines the strengths of traditional hand-crafted controllers and model-free deep reinforcement learning (RL). BCF thrives in the robotics domain, where reliable …
View article: Human-Timescale Adaptation in an Open-Ended Task Space
Human-Timescale Adaptation in an Open-Ended Task Space Open
Foundation models have shown impressive adaptation and scalability in supervised and self-supervised learning problems, but so far these successes have not fully translated to reinforcement learning (RL). In this work, we demonstrate that …
View article: Developing, Evaluating and Scaling Learning Agents in Multi-Agent Environments
Developing, Evaluating and Scaling Learning Agents in Multi-Agent Environments Open
The Game Theory & Multi-Agent team at DeepMind studies several aspects of multi-agent learning ranging from computing approximations to fundamental concepts in game theory to simulating social dilemmas in rich spatial environments and trai…
View article: The Challenges of Exploration for Offline Reinforcement Learning
The Challenges of Exploration for Offline Reinforcement Learning Open
Offline Reinforcement Learning (ORL) enablesus to separately study the two interlinked processes of reinforcement learning: collecting informative experience and inferring optimal behaviour. The second step has been widely studied in the o…
View article: Efficient and stable reinforcement learning for robotics
Efficient and stable reinforcement learning for robotics Open
Reinforcement Learning (RL) has long been used for learning behaviour through agent-collected experience, recently boosted by deep neural networks. However, typical deep RL agents require millions of training data samples, equating to days…
View article: Zero-Shot Uncertainty-Aware Deployment of Simulation Trained Policies on Real-World Robots
Zero-Shot Uncertainty-Aware Deployment of Simulation Trained Policies on Real-World Robots Open
While deep reinforcement learning (RL) agents have demonstrated incredible potential in attaining dexterous behaviours for robotics, they tend to make errors when deployed in the real world due to mismatches between the training and execut…
View article: Is Curiosity All You Need? On the Utility of Emergent Behaviours from Curious Exploration
Is Curiosity All You Need? On the Utility of Emergent Behaviours from Curious Exploration Open
Curiosity-based reward schemes can present powerful exploration mechanisms which facilitate the discovery of solutions for complex, sparse or long-horizon tasks. However, as the agent learns to reach previously unexplored spaces and the ob…
View article: Bayesian Controller Fusion: Leveraging Control Priors in Deep Reinforcement Learning for Robotics
Bayesian Controller Fusion: Leveraging Control Priors in Deep Reinforcement Learning for Robotics Open
We present Bayesian Controller Fusion (BCF): a hybrid control strategy that combines the strengths of traditional hand-crafted controllers and model-free deep reinforcement learning (RL). BCF thrives in the robotics domain, where reliable …
View article: Multiplicative Controller Fusion: Leveraging Algorithmic Priors for Sample-efficient Reinforcement Learning and Safe Sim-To-Real Transfer
Multiplicative Controller Fusion: Leveraging Algorithmic Priors for Sample-efficient Reinforcement Learning and Safe Sim-To-Real Transfer Open
Learning-based approaches often outperform hand-coded algorithmic solutions for many problems in robotics. However, learning long-horizon tasks on real robot hardware can be intractable, and transferring a learned policy from simulation to…
View article: Learning Arbitrary-Goal Fabric Folding with One Hour of Real Robot Experience
Learning Arbitrary-Goal Fabric Folding with One Hour of Real Robot Experience Open
Manipulating deformable objects, such as fabric, is a long standing problem in robotics, with state estimation and control posing a significant challenge for traditional methods. In this paper, we show that it is possible to learn fabric f…
View article: Residual Reactive Navigation: Combining Classical and Learned Navigation Strategies For Deployment in Unknown Environments
Residual Reactive Navigation: Combining Classical and Learned Navigation Strategies For Deployment in Unknown Environments Open
In this work we focus on improving the efficiency and generalisation of learned navigation strategies when transferred from its training environment to previously unseen ones. We present an extension of the residual reinforcement learning …
View article: Multiplicative Controller Fusion: Leveraging Algorithmic Priors for\n Sample-efficient Reinforcement Learning and Safe Sim-To-Real Transfer
Multiplicative Controller Fusion: Leveraging Algorithmic Priors for\n Sample-efficient Reinforcement Learning and Safe Sim-To-Real Transfer Open
Learning-based approaches often outperform hand-coded algorithmic solutions\nfor many problems in robotics. However, learning long-horizon tasks on real\nrobot hardware can be intractable, and transferring a learned policy from\nsimulation…
View article: Evaluating task-agnostic exploration for fixed-batch learning of arbitrary future tasks
Evaluating task-agnostic exploration for fixed-batch learning of arbitrary future tasks Open
Deep reinforcement learning has been shown to solve challenging tasks where large amounts of training experience is available, usually obtained online while learning the task. Robotics is a significant potential application domain for many…
View article: Ctrl-Z: Recovering from Instability in Reinforcement Learning
Ctrl-Z: Recovering from Instability in Reinforcement Learning Open
When learning behavior, training data is often generated by the learner itself; this can result in unstable training dynamics, and this problem has particularly important applications in safety-sensitive real-world control tasks such as ro…
View article: Sim-to-Real Transfer of Robot Learning with Variable Length Inputs
Sim-to-Real Transfer of Robot Learning with Variable Length Inputs Open
Current end-to-end deep Reinforcement Learning (RL) approaches require jointly learning perception, decision-making and low-level control from very sparse reward signals and high-dimensional inputs, with little capability of incorporating …
View article: Zero-shot Sim-to-Real Transfer with Modular Priors.
Zero-shot Sim-to-Real Transfer with Modular Priors. Open
Current end-to-end deep Reinforcement Learning (RL) approaches require jointly learning perception, decision-making and low-level control from very sparse reward signals and high-dimensional inputs, with little capability of incorporating …