Search Results for author:
Found 19 papers, 5 papers with code
The Cross-environment Hyperparameter Setting Benchmark for Reinforcement Learning
This paper introduces a new empirical methodology, the Cross-environment Hyperparameter Setting Benchmark, that compares RL algorithms across environments using a single hyperparameter setting, encouraging algorithmic development which is insensitive to hyperparameters.
Investigating the Interplay of Prioritized Replay and Generalization
One exception is Prioritized Experience Replay (PER), where sampling is done proportionally to TD errors, inspired by the success of prioritized sweeping in dynamic programming.
When is Offline Policy Selection Sample Efficient for Reinforcement Learning?
As a result, no OPS method can be more sample efficient than OPE in the worst case.
Empirical Design in Reinforcement Learning
The objective of this document is to provide answers on how we can use our unprecedented compute to do good science in reinforcement learning, as well as stay alert to potential pitfalls in our empirical design.
Robust Losses for Learning Value Functions
We start from a formalization of robust losses, then derive sound gradient-based approaches to minimize these losses in both the online off-policy prediction and control settings.
A Temporal-Difference Approach to Policy Gradient Estimation
The policy gradient theorem (Sutton et al., 2000) prescribes the usage of a cumulative discounted state distribution under the target policy to approximate the gradient.
A Generalized Projected Bellman Error for Off-policy Value Estimation in Reinforcement Learning
Many algorithms have been developed for off-policy value estimation based on the linear mean squared projected Bellman error (MSPBE) and are sound under linear function approximation.
Contraction $\mathcal{L}_1$-Adaptive Control using Gaussian Processes
We present $\mathcal{CL}_1$-$\mathcal{GP}$, a control framework that enables safe simultaneous learning and control for systems subject to uncertainties.
Gradient Temporal-Difference Learning with Regularized Corrections
It is still common to use Q-learning and temporal difference (TD) learning-even though they have divergence issues and sound Gradient TD alternatives exist-because divergence seems rare and they typically perform well.
L1-GP: L1 Adaptive Control with Bayesian Learning
We present L1-GP, an architecture based on L1 adaptive control and Gaussian Process Regression (GPR) for safe simultaneous control and learning.
Learning Probabilistic Intersection Traffic Models for Trajectory Prediction
The safety of these agents is dependent on their ability to predict collisions with other vehicles' future trajectories for replanning and collision avoidance.
Learning Macroscopic Brain Connectomes via Group-Sparse Factorization
We develop an efficient optimization strategy for this extremely high-dimensional sparse problem, by reducing the number of parameters using a greedy algorithm designed specifically for the problem.
Intent-Aware Probabilistic Trajectory Estimation for Collision Prediction with Uncertainty Quantification
We show that the uncertainty region for obstacle positions can be expressed in terms of a combination of polynomials generated with Gaussian process regression.
Proximity Queries for Absolutely Continuous Parametric Curves
In motion planning problems for autonomous robots, such as self-driving cars, the robot must ensure that its planned path is not in close proximity to obstacles in the environment.
Robotics Computational Geometry Graphics
Supervised autoencoders: Improving generalization performance with unsupervised regularizers
A common strategy to improve generalization has been through the use of regularizers, typically as a norm constraining the parameters.
Online Off-policy Prediction
The ability to learn behavior-contingent predictions online and off-policy has long been advocated as a key capability of predictive-knowledge learning systems but remained an open algorithmic challenge for decades.
General Value Function Networks
A general purpose strategy for state construction is to learn the state update using a Recurrent Neural Network (RNN), which updates the internal state using the current internal state and the most recent observation.
Organizing Experience: A Deeper Look at Replay Mechanisms for Sample-based Planning in Continuous State Domains
We show that a model, as opposed to a replay buffer, is particularly useful for specifying which states to sample from during planning, such as predecessor states that propagate information in reverse from a state more quickly.
Discovery of Predictive Representations With a Network of General Value Functions
We investigate a framework for discovery: curating a large collection of predictions, which are used to construct the agent's representation of the world.
