arXiv research

DPN combines model-based and model-free reinforcement learning for efficient planning.

Model-free RL agents can learn effective planning without explicit model.

AOP combines model-based planning with model-free learning to handle lifelong learning challenges.

End-to-end Dyna-AIL combines planning and learning for faster convergence.

The paper tackles model-based RL's inaccuracy issue by dynamically adjusting planning horizons.

Unified reinforcement learning methods using hybrid inference.

Proposes learning latent reward model for planning from rewards.

PackIt creates a virtual space for testing geometric planning skills.

Efficient Q-learning for metric spaces without planning oracles.

DADS discovers skills with predictable outcomes from unlabeled data.

A method uses RL to learn abstractions for planning, improving robot navigation and manipulation tasks.

Model-Based Offline Planning (MBOP) learns models from offline data to control systems directly.

Unified Latent Dynamics unifies model-free and model-based reinforcement learning.

This paper presents a unifying framework for reinforcement learning and planning.

XLVINs improve data efficiency in implicit planning by leveraging latent space.

A key challenge in complex visuomotor control is learning abstract representations that are effective for specifying goals, planning, and generalization. To this end, we introduce universal planning networks (UPN). UPNs embed differentiable planning within a goal-directed policy. This planning computation unrolls a for…

The study compares reinforcement learning models and finds model-based approaches superior for complex MDPs.

PlanGAN uses GANs to plan efficient trajectories for multi-goal tasks in sparse reward environments.

Planning has been very successful for control tasks with known environment dynamics. To leverage planning in unknown environments, the agent needs to learn the dynamics from interactions with the world. However, learning dynamics models that are accurate enough for planning has been a long-standing challenge, especiall…

We introduce Imagination-Augmented Agents (I2As), a novel architecture for deep reinforcement learning combining model-free and model-based aspects. In contrast to most existing model-based reinforcement learning and planning methods, which prescribe how a model should be used to arrive at a policy, I2As learn to inter…

This paper introduces the QMDP-net, a neural network architecture for planning under partial observability. The QMDP-net combines the strengths of model-free learning and model-based planning. It is a recurrent policy network, but it represents a policy for a parameterized set of tasks by connecting a model with a plan…

Hierarchical Foresight improves robot vision tasks by planning long-term goals.

RandQL is a new model-free algorithm for MDPs with a novel learning rate randomization approach.

Residual algorithms improve reinforcement learning performance.

CriticSMC improves planning efficiency in constrained environments.

Building deep reinforcement learning agents that can generalize and adapt to unseen environments remains a fundamental challenge for AI. This paper describes progresses on this challenge in the context of man-made environments, which are visually diverse but contain intrinsic semantic regularities. We propose a hybrid …

Proposes a method for model-based RL in complex environments without perfect simulators.

NIPA aims to translate brain learning mechanisms into scalable Bayesian inference.

In this paper we apply change of numeraire techniques to the optimal transport approach for computing model-free prices of derivatives in a two periods model. In particular, we consider the optimal transport plan constructed in \cite{HobsonKlimmek2013} as well as the one introduced in \cite{BeiglJuil} and further studi…

SAVE combines Q-learning and MCTS with amortized value estimates for improved performance.

Develops a method to plan exploration that learns strong policies with fewer samples.

Combines MFRL and MPC for data-efficient high performance in robotics.

Reinforcement learning optimizes robot trajectories for unknown dynamics.

A new approach for deep exploration in sparse reward reinforcement learning.

Framework uses RL and simulation for optimal microgrid energy storage planning.

Consider mutli-goal tasks that involve static environments and dynamic goals. Examples of such tasks, such as goal-directed navigation and pick-and-place in robotics, abound. Two types of Reinforcement Learning (RL) algorithms are used for such tasks: model-free or model-based. Each of these approaches has limitations.…

In the quest for efficient and robust reinforcement learning methods, both model-free and model-based approaches offer advantages. In this paper we propose a new way of explicitly bridging both approaches via a shared low-dimensional learned encoding of the environment, meant to capture summarizing abstractions. We sho…

XLVINs improve deep reinforcement learning by combining self-supervised learning and neural algorithmic reasoning.

Reinforcement learning and planning methods require an objective or reward function that encodes the desired behavior. Yet, in practice, there is a wide range of scenarios where an objective is difficult to provide programmatically, such as tasks with visual observations involving unknown object positions or deformable…

Combining deep model-free reinforcement learning with on-line planning is a promising approach to building on the successes of deep RL. On-line planning with look-ahead trees has proven successful in environments where transition models are known a priori. However, in complex environments where transition models need t…

Paper develops efficient RL algorithm for general value function approximation.

Maximize to Explore integrates RL components for efficient policy discovery.

Paper benchmarks MBRL algorithms with new environments.

We adapt the ideas underlying the success of Deep Q-Learning to the continuous action domain. We present an actor-critic, model-free algorithm based on the deterministic policy gradient that can operate over continuous action spaces. Using the same learning algorithm, network architecture and hyper-parameters, our algo…

New RL method uses distance between states instead of rewards for sparse reward environments.

Study shows plug-in model-based reinforcement learning is minimax optimal.

We introduce an approach for deep reinforcement learning (RL) that improves upon the efficiency, generalization capacity, and interpretability of conventional approaches through structured perception and relational reasoning. It uses self-attention to iteratively reason about the relations between entities in a scene a…

Deep reinforcement learning has been successfully applied to several visual-input tasks using model-free methods. In this paper, we propose a model-based approach that combines learning a DNN-based transition model with Monte Carlo tree search to solve a block-placing task in Minecraft. Our learned transition model pre…