Search Results for author:
Found 39 papers, 15 papers with code
Function-space Parameterization of Neural Networks for Sequential Learning
Our parameterization offers: (i) a way to scale function-space methods to large data sets via sparsification, (ii) retention of prior knowledge when access to past data is limited, and (iii) a mechanism to incorporate new data without retraining.
AgentMixer: Multi-Agent Correlated Policy Factorization
Inspired by the concept of correlated equilibrium, we propose to introduce a \textit{strategy modification} to provide a mechanism for agents to correlate their policies.
Optimistic Multi-Agent Policy Gradient for Cooperative Tasks
However, with function approximation optimism can amplify overestimation and thus fail on complex tasks.
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.
On the Benefit of Optimal Transport for Curriculum Reinforcement Learning
In this work, we focus on framing curricula as interpolations between task distributions, which has previously been shown to be a viable approach to CRL.
Tracking Control for a Spherical Pendulum via Curriculum Reinforcement Learning
Reinforcement Learning (RL) allows learning non-trivial robot control laws purely from data.
Monte-Carlo tree search with uncertainty propagation via optimal transport
We introduce a novel backup operator that computes value nodes as the Wasserstein barycenter of their action-value children nodes; thus, propagating the uncertainty of the estimate across the tree to the root node.
Sparse Function-space Representation of Neural Networks
Deep neural networks (NNs) are known to lack uncertainty estimates and struggle to incorporate new data.
Towards Energy Efficient Control for Commercial Heavy-Duty Mobile Cranes: Modeling Hydraulic Pressures using Machine Learning
A sizable part of the fleet of heavy-duty machinery in the construction equipment industry uses the conventional valve-controlled load-sensing hydraulics.
Simplified Temporal Consistency Reinforcement Learning
This applies when using pure planning with a dynamics model conditioned on the representation, but, also when utilizing the representation as policy and value function features in model-free RL.
Swapped goal-conditioned offline reinforcement learning
Offline goal-conditioned reinforcement learning (GCRL) can be challenging due to overfitting to the given dataset.
Prioritized offline Goal-swapping Experience Replay
In goal-conditioned offline reinforcement learning, an agent learns from previously collected data to go to an arbitrary goal.
Hierarchical Imitation Learning with Vector Quantized Models
The ability to plan actions on multiple levels of abstraction enables intelligent agents to solve complex tasks effectively.
Redeeming Intrinsic Rewards via Constrained Optimization
However, on easy exploration tasks, the agent gets distracted by intrinsic rewards and performs unnecessary exploration even when sufficient task (also called extrinsic) reward is available.
Adaptive Behavior Cloning Regularization for Stable Offline-to-Online Reinforcement Learning
Offline reinforcement learning, by learning from a fixed dataset, makes it possible to learn agent behaviors without interacting with the environment.
Continuous Monte Carlo Graph Search
To address this problem, we propose Continuous Monte Carlo Graph Search (CMCGS), an extension of MCTS to online planning in environments with continuous state and action spaces.
Partially Observable Markov Decision Processes in Robotics: A Survey
Noisy sensing, imperfect control, and environment changes are defining characteristics of many real-world robot tasks.
Learning Progress Driven Multi-Agent Curriculum
Inspired by the success of CRL in single-agent settings, a few works have attempted to apply CRL to multi-agent reinforcement learning (MARL) using the number of agents to control task difficulty.
Multi-agent Reinforcement Learning
Open-Ended Question Answering
+3
Topological Experience Replay
State-of-the-art deep Q-learning methods update Q-values using state transition tuples sampled from the experience replay buffer.
GPU-Accelerated Policy Optimization via Batch Automatic Differentiation of Gaussian Processes for Real-World Control
The ability of Gaussian processes (GPs) to predict the behavior of dynamical systems as a more sample-efficient alternative to parametric models seems promising for real-world robotics research.
A Unified Perspective on Value Backup and Exploration in Monte-Carlo Tree Search
In this work, we propose two methods for improving the convergence rate and exploration based on a newly introduced backup operator and entropy regularization.
State-Conditioned Adversarial Subgoal Generation
Hierarchical reinforcement learning (HRL) proposes to solve difficult tasks by performing decision-making and control at successively higher levels of temporal abstraction.
Metrics Matter: A Closer Look on Self-Paced Reinforcement Learning
Experiments demonstrate that the resulting introduction of metric structure into the curriculum allows for a well-behaving non-parametric version of SPRL that leads to stable learning performance across tasks.
Boosted Curriculum Reinforcement Learning
This approach, which we refer to as boosted curriculum reinforcement learning (BCRL), has the benefit of naturally increasing the representativeness of the functional space by adding a new residual each time a new task is presented.
Reinforcement Learning using Guided Observability
Due to recent breakthroughs, reinforcement learning (RL) has demonstrated impressive performance in challenging sequential decision-making problems.
Neural Network Controller for Autonomous Pile Loading Revised
We have recently proposed two pile loading controllers that learn from human demonstrations: a neural network (NNet) [1] and a random forest (RF) controller [2].
A Probabilistic Interpretation of Self-Paced Learning with Applications to Reinforcement Learning
Across machine learning, the use of curricula has shown strong empirical potential to improve learning from data by avoiding local optima of training objectives.
Technical Report: The Policy Graph Improvement Algorithm
The policy graph improvement (PGI) algorithm for POMDPs represents the policy as a fixed size policy graph and improves the policy monotonically.
Multi-Sensor Next-Best-View Planning as Matroid-Constrained Submodular Maximization
We consider the problem of creating a 3D model using depth images captured by a team of multiple robots.
Convex Regularization in Monte-Carlo Tree Search
Monte-Carlo planning and Reinforcement Learning (RL) are essential to sequential decision making.
Machine Learning Based Mobile Network Throughput Classification
Identifying mobile network problems in 4G cells is more challenging when the complexity of the network increases, and privacy concerns limit the information content of the data.
Self-Paced Deep Reinforcement Learning
Curriculum reinforcement learning (CRL) improves the learning speed and stability of an agent by exposing it to a tailored series of tasks throughout learning.
Deep Adversarial Reinforcement Learning for Object Disentangling
The ARL framework utilizes an adversary, which is trained to steer the original agent, the protagonist, to challenging states.
Long-Term Visitation Value for Deep Exploration in Sparse Reward Reinforcement Learning
Empirical results on classic and novel benchmarks show that the proposed approach outperforms existing methods in environments with sparse rewards, especially in the presence of rewards that create suboptimal modes of the objective function.
Generalized Mean Estimation in Monte-Carlo Tree Search
Finally, we empirically demonstrate the effectiveness of our method in well-known MDP and POMDP benchmarks, showing significant improvement in performance and convergence speed w. r. t.
Model-based Lookahead Reinforcement Learning
Model-based Reinforcement Learning (MBRL) allows data-efficient learning which is required in real world applications such as robotics.
Information Gathering in Decentralized POMDPs by Policy Graph Improvement
Decentralized policies for information gathering are required when multiple autonomous agents are deployed to collect data about a phenomenon of interest without the ability to communicate.
Compatible Natural Gradient Policy Search
Trust-region methods have yielded state-of-the-art results in policy search.
An Algorithmic Perspective on Imitation Learning
This process of learning from demonstrations, and the study of algorithms to do so, is called imitation learning.
