Search Results for author:
Found 37 papers, 19 papers with code
Planning to Explore via Latent Disagreement
To solve complex tasks, intelligent agents first need to explore their environments.
Learning to Model the World with Language
To interact with humans in the world, agents need to understand the diverse types of language that people use, relate them to the visual world, and act based on them.
Video Prediction Models as Rewards for Reinforcement Learning
A promising approach is to extract preferences for behaviors from unlabeled videos, which are widely available on the internet.
Mastering Diverse Domains through World Models
General intelligence requires solving tasks across many domains.
Evaluating Long-Term Memory in 3D Mazes
However, most benchmark tasks in reinforcement learning do not test long-term memory in agents, slowing down progress in this important research direction.
Learning Robust Dynamics through Variational Sparse Gating
Learning world models from their sensory inputs enables agents to plan for actions by imagining their future outcomes.
Temporally Consistent Transformers for Video Generation
To generate accurate videos, algorithms have to understand the spatial and temporal dependencies in the world.
Masked World Models for Visual Control
Yet the current approaches typically train a single model end-to-end for learning both visual representations and dynamics, making it difficult to accurately model the interaction between robots and small objects.
Model-based Reinforcement Learning
Reinforcement Learning (RL)
+1
DayDreamer: World Models for Physical Robot Learning
Learning a world model to predict the outcomes of potential actions enables planning in imagination, reducing the amount of trial and error needed in the real environment.
Deep Hierarchical Planning from Pixels
Despite operating in latent space, the decisions are interpretable because the world model can decode goals into images for visualization.
Information is Power: Intrinsic Control via Information Capture
We study this question in dynamic partially-observed environments, and argue that a compact and general learning objective is to minimize the entropy of the agent's state visitation estimated using a latent state-space model.
Discovering and Achieving Goals via World Models
How can artificial agents learn to solve many diverse tasks in complex visual environments in the absence of any supervision?
Benchmarking the Spectrum of Agent Capabilities
We hope that Crafter will accelerate research progress by quickly evaluating a wide spectrum of abilities.
Intrinsic Control of Variational Beliefs in Dynamic Partially-Observed Visual Environments
We study this question in dynamic partially-observed environments, and argue that a compact and general learning objective is to minimize the entropy of the agent's state visitation estimated using a latent state-space model.
Discovering and Achieving Goals with World Models
How can an artificial agent learn to solve a wide range of tasks in a complex visual environment in the absence of external supervision?
Clockwork Variational Autoencoders
We introduce the Clockwork VAE (CW-VAE), a video prediction model that leverages a hierarchy of latent sequences, where higher levels tick at slower intervals.
Video Prediction with Variational Temporal Hierarchies
Deep learning has shown promise for accurately predicting high-dimensional video sequences.
On Trade-offs of Image Prediction in Visual Model-Based Reinforcement Learning
In this paper, we study a number of design decisions for the predictive model in visual MBRL algorithms, focusing specifically on methods that use a predictive model for planning.
Model-based Reinforcement Learning
reinforcement-learning
+1
Evaluating Agents without Rewards
Moreover, input entropy and information gain correlate more strongly with human similarity than task reward does, suggesting the use of intrinsic objectives for designing agents that behave similarly to human players.
Models, Pixels, and Rewards: Evaluating Design Trade-offs in Visual Model-Based Reinforcement Learning
In this paper, we study a number of design decisions for the predictive model in visual MBRL algorithms, focusing specifically on methods that use a predictive model for planning.
Model-based Reinforcement Learning
Reinforcement Learning (RL)
Latent Skill Planning for Exploration and Transfer
To quickly solve new tasks in complex environments, intelligent agents need to build up reusable knowledge.
Mastering Atari with Discrete World Models
The world model uses discrete representations and is trained separately from the policy.
Ranked #3 on
Atari Games
on Atari 2600 Skiing
(using extra training data)
Action and Perception as Divergence Minimization
While the narrow objectives correspond to domain-specific rewards as typical in reinforcement learning, the general objectives maximize information with the environment through latent variable models of input sequences.
Sophisticated Inference
In this paper, we consider a sophisticated kind of active inference, using a recursive form of expected free energy.
Planning to Explore via Self-Supervised World Models
Reinforcement learning allows solving complex tasks, however, the learning tends to be task-specific and the sample efficiency remains a challenge.
Dream to Control: Learning Behaviors by Latent Imagination
Learned world models summarize an agent's experience to facilitate learning complex behaviors.
Reliable Uncertainty Estimates in Deep Neural Networks using Noise Contrastive Priors
NCPs are compatible with any model that can output uncertainty estimates, are easy to scale, and yield reliable uncertainty estimates throughout training.
Bayesian Layers: A Module for Neural Network Uncertainty
We describe Bayesian Layers, a module designed for fast experimentation with neural network uncertainty.
Modulated Policy Hierarchies
To achieve this, we study different modulation signals and exploration for hierarchical controllers.
Learning Latent Dynamics for Planning from Pixels
Planning has been very successful for control tasks with known environment dynamics.
Ranked #2 on
Continuous Control
on DeepMind Cheetah Run (Images)
Noise Contrastive Priors for Functional Uncertainty
NCPs are compatible with any model that can output uncertainty estimates, are easy to scale, and yield reliable uncertainty estimates throughout training.
Sample-Efficient Reinforcement Learning with Stochastic Ensemble Value Expansion
Integrating model-free and model-based approaches in reinforcement learning has the potential to achieve the high performance of model-free algorithms with low sample complexity.
Sim-to-Real: Learning Agile Locomotion For Quadruped Robots
The control policies are learned in a physics simulator and then deployed on real robots.
Generative Interest Estimation for Document Recommendations
Then, we capture a user's interest as a generative model in the space of the document representations.
TensorFlow Agents: Efficient Batched Reinforcement Learning in TensorFlow
We introduce TensorFlow Agents, an efficient infrastructure paradigm for building parallel reinforcement learning algorithms in TensorFlow.
Learning Hierarchical Information Flow with Recurrent Neural Modules
We propose ThalNet, a deep learning model inspired by neocortical communication via the thalamus.
Deep Reinforcement Learning From Raw Pixels in Doom
Using current reinforcement learning methods, it has recently become possible to learn to play unknown 3D games from raw pixels.
