On-device research index

arXiv research

A locally-built, LLM-digested index of recent arXiv papers in quant finance, geometry/topology, and statistical ML — keyword search served straight from SQLite on this machine.

168,695 papers · 148 categories

Trend · papers per month

count % of corpus

8 results for “MuZero”

Equivariant MuZerocs.LG

Equivariant MuZero improves generalization in procedurally generated environments.

problem Leveraging world models for robust generalization in real-world scenarios.

method Incorporates symmetries of the environment into MuZero's world-model architecture, making MuZero equivariant to certain groups of transformations.

result Improves data efficiency and generalization capabilities, demonstrated on procedurally generated MiniPacman and Chaser.

reinforcement learning model-based equivariance world model generalization

2023-02-09abs ↗pdf ↗

Visualizing MuZero Modelscs.LG

MuZero visualizes its internal representations to stabilize planning.

problem Stability issues in MuZero's planning process.

method Visualized MuZero's latent representations and proposed regularization techniques.

result Action trajectories diverge between observation embeddings and internal state transitions, leading to instability.

MuZero reinforcement learning value equivalent models latent representations planning stability

2021-02-25abs ↗pdf ↗

Mastering Atari, Go, Chess and Shogi by Planning with a Learned Modelcs.LG

Constructing agents with planning capabilities has long been one of the main challenges in the pursuit of artificial intelligence. Tree-based planning methods have enjoyed huge success in challenging domains, such as chess and Go, where a perfect simulator is available. However, in real-world problems the dynamics gove…

2019-11-19abs ↗pdf ↗

A Note on Loss Functions and Error Compounding in Model-based Reinforcement Learningcs.LG

Clarifies model-based RL's theoretical issues and counterexamples for popular losses.

problem Model-based reinforcement learning's empirical performance vs. theoretical properties and popular loss functions.

method Analyzes empirical and theoretical aspects of model-based RL and constructs counterexamples for losses.

result MuZero loss fails in stochastic and deterministic environments, leading to exponential sample complexity.

reinforcement learning model-based loss functions error compounding MuZero

2024-04-15abs ↗pdf ↗

Vector Quantized Models for Planningcs.LG

New approach handles stochastic and partially-observable environments using discrete autoencoders and Monte Carlo tree search.

problem Challenges in planning for stochastic and partially-observable environments.

method Uses discrete autoencoders and a stochastic variant of Monte Carlo tree search.

result Significantly outperforms MuZero on stochastic chess and scales to DeepMind Lab.

model-based RL stochastic environments Monte Carlo tree search discrete autoencoders planning

2021-06-08abs ↗pdf ↗

Cost-Driven Representation Learning for Linear Quadratic Gaussian Control: Part IIcs.LG

Study cost-driven state representation learning for control from partial observations.

problem Learning state representation for control from partial and high-dimensional observations.

method Cost-driven state representation learning via predicting cumulative costs.

result Established finite-sample guarantees for near-optimal representation and controller.

Control Representation Learning Cumulative Costs Linear Quadratic Gaussian Finite-Sample Guarantees

2026-03-08abs ↗pdf ↗

The LoCA Regret: A Consistent Metric to Evaluate Model-Based Behavior in Reinforcement Learningcs.LG

Introduces LoCA regret to evaluate model-based RL methods.

problem Lack of consistent metrics to evaluate model-based RL methods.

method Inspired by neuroscience, introduces LoCA regret to measure model-based behavior.

result LoCA regret can identify model-based behavior and assess how close methods are to optimal model-based behavior.

Reinforcement Learning Model-Based Behavior Evaluation LoCA Regret Neuroscience

2020-07-07abs ↗pdf ↗

Cost-Driven Representation Learning for Linear Quadratic Gaussian Control: Part Ics.LG

Study learns state representations from observations for control, proving guarantees.

problem Learning state representations from high-dimensional observations for control.

method Cost-driven approach, learning latent state model to predict costs.

result Proves finite-sample guarantees for near-optimal state representation and controller.

Representation Learning Control Theory Linear Quadratic Gaussian Cost-Driven Finite-Sample Guarantees

2022-12-30abs ↗pdf ↗