Synthesizing Neural Network Controllers with Probabilistic Model-Based Reinforcement Learning
Explore this paper's citation graph
Summary
An algorithm for rapidly learning neural network policies for robotics systems that follows the model-based reinforcement learning paradigm and improves upon existing algorithms: PILeO and a sample-based version of PILeo with neural network dynamics (Deep-PILeO).
- Type
- preprint
- Published
- 2018-03-06
- Cited by
- 41
- References
- 32
- Access
- Open access
- OpenAlex
- https://openalex.org/W2790331160
- Semantic Scholar
- https://api.semanticscholar.org/CorpusID:3730440
Keywords
Reinforcement learning, Computer science, Artificial neural network, Artificial intelligence, Benchmark (surveying)
References
- Iterative Linear Quadratic Regulator Design for Nonlinear Biological Movement Systems
- Differential dynamic programming
- Model learning for robot control: a survey
- Learning legged swimming gaits from experience
- Dynamics and trajectory optimization for a soft spatial fluidic elastomer manipulator
- Locally Weighted Learning for Control
- PEGASUS: A policy search method for large MDPs and POMDPs
- On the difficulty of training recurrent neural networks
- Control-limited differential dynamic programming
- Gaussian Processes for Data-Efficient Learning in Robotics and Control
- Dropout: a simple way to prevent neural networks from overfitting
- A comparison of direct and model-based reinforcement learning
- Learning Neural Network Policies with Guided Policy Search under Unknown Dynamics
- Probabilistic Differential Dynamic Programming
- Simulation-Based Optimization with Stochastic Approximation Using Common Random Numbers
- Weight Uncertainty in Neural Network
- Sparse Spectrum Gaussian Process Regression
- Efficient reinforcement learning using Gaussian processes
- Black-box data-efficient policy search for robotics
- GP-ILQG: Data-driven Robust Optimal Control for Uncertain Nonlinear Dynamical Systems
Cited by
- A Survey on Policy Search Algorithms for Learning Robot Controllers in a Handful of Trials
- Learning Latent Dynamics for Planning from Pixels
- Safer reinforcement learning for robotics
- Uncertainty Aware Learning from Demonstrations in Multiple Contexts using Bayesian Neural Networks
- Where Off-Policy Deep Reinforcement Learning Fails
- Model-based reinforcement learning: A survey
- Model-Based Bayesian Sparse Sampling for Data Efficient Control
- Micro-Data Reinforcement Learning for Adaptive Robots. (Apprentissage micro-data pour l'adaptation en robotique)
- Unifying Variational Inference and PAC-Bayes for Supervised Learning that Scales
- A selected review on reinforcement learning based control for autonomous underwater vehicles
- Self-Supervised Object-Level Deep Reinforcement Learning
- Deep vs. Deep Bayesian: Reinforcement Learning on a Multi-Robot Competitive Experiment
- Robot Action Selection Learning via Layered Dimension Informed Program Synthesis
- Relevance-Guided Modeling of Object Dynamics for Reinforcement Learning
- Mastering Atari with Discrete World Models
- Data-Efficient Robot Learning using Priors from Simulators. (Apprentissage efficace en données pour la robotique à l'aide de simulateurs)
- Multimodal dynamics modeling for off-road autonomous vehicles
- Deep vs. Deep Bayesian: Faster Reinforcement Learning on a Multi-robot Competitive Experiment
- A Survey of Sim-to-Real Transfer Techniques Applied to Reinforcement Learning for Bioinspired Robots
- Traded Control of Human–Machine Systems for Sequential Decision-Making Based on Reinforcement Learning
Related papers
- Online learning of neural networks based on a model-free control algorithm
- Artificial neural networks - ICANN 2010 : 20th International Conference, Thessaloniki, Greece, September 15-18, 2010 : proceedings
- application of Neural Networks to Adaptive Control of Nonlinear Systems
- Fuzzy entropy based combined learning algorithm for neural networks
- Artificial Neural Networks and Machine Learning – ICANN 2019: Deep Learning 28th International Conference on Artificial Neural Networks, Munich, Germany, September 17–19, 2019, Proceedings, Part II
- A neural network approach to adaptive state-space partitioning