A new method improves actor-critic RL by integrating HMC, enhancing policy distribution and exploration.
problem Actor-critic RL yields suboptimal policies due to amortization gap and insufficient exploration.
method Integrating Hamiltonian Monte Carlo (HMC) into the actor-critic RL framework.
result Improves policy distribution and exploration, leading to better policy estimates and higher returns.
AACC improves RL performance in changing environments.
problem Deterioration of RL performance in real-world tasks.
method Formalizes CMDPs, proposes AACC for contextual RL.
result AACC outperforms baselines in various simulated environments.
Study efficient convergence of RL algorithm with function approximation.
problem Convergence of actor-critic algorithm with nonlinear function approximation.
method Stochastic gradient descent ascent with adaptive proximal term, Polyak-Łojasiewicz condition.
result First efficient convergence result with rate of O(sqrt{ln(N d G^2) / N}).
New actor-critic algorithm achieves optimal sample efficiency in RL.
problem Achieving ε-optimal policies with minimal samples in RL. method Integrates optimism, off-policy critic estimation, and rare-switching policy resets.
result Sample complexity of O(dH5log∣A∣/ε2+dH4log∣F∣/ε2) trajectories. Both generative adversarial networks (GAN) in unsupervised learning and actor-critic methods in reinforcement learning (RL) have gained a reputation for being difficult to optimize. Practitioners in both fields have amassed a large number of strategies to mitigate these instabilities and improve training. Here we show …
Model-free deep reinforcement learning (RL) algorithms have been demonstrated on a range of challenging decision making and control tasks. However, these methods typically suffer from two major challenges: very high sample complexity and brittle convergence properties, which necessitate meticulous hyperparameter tuning…
Applying probabilistic models to reinforcement learning (RL) enables the application of powerful optimisation tools such as variational inference to RL. However, existing inference frameworks and their algorithms pose significant challenges for learning optimal policies, e.g., the absence of mode capturing behaviour in…
Deep Reinforcement Learning (DRL) algorithms for continuous action spaces are known to be brittle toward hyperparameters as well as \cut{being}sample inefficient. Soft Actor Critic (SAC) proposes an off-policy deep actor critic algorithm within the maximum entropy RL framework which offers greater stability and empiric…
New PAC-Bayesian approach stabilizes actor-critic learning.
problem Training instability in actor-critic algorithms.
method Employing PAC-Bayesian bound as the critic training objective.
result Significant improvement in online learning performance.
RL agents outperform baselines in asset allocation.
problem Optimizing asset allocation using reinforcement learning.
method Model-free deep RL agents trained on real-world stock prices.
result RL agents significantly outperformed random and uniform allocation.
New RL method MAC improves performance in sparse reward settings.
problem Slow mixing in large state spaces or sparse rewards.
method Multi-level Monte Carlo Actor-Critic (MAC) algorithm.
result Achieves convergence rate comparable to state-of-the-art AC algorithms.
ESAC combines genetic methods with RL to improve scalability and efficiency.
problem Combining genetic scalability with RL's data efficiency and optimal control.
method Combines Evolution Strategies (ES) with Soft Actor-Critic (SAC) to enable skill transfer and reduce hyperparameter sensitivity.
result Demonstrates improved performance and sample efficiency in challenging tasks.
D2D-SPL uses discrete states and a classifier to train RL faster.
problem Training neural networks in RL due to correlated samples.
method Discretizes state space, uses actor-critic, selects input/target pairs, trains classifier.
result Trains faster than state-of-the-art methods.
We propose a new objective, the counterfactual objective, unifying existing objectives for off-policy policy gradient algorithms in the continuing reinforcement learning (RL) setting. Compared to the commonly used excursion objective, which can be misleading about the performance of the target policy when deployed, our…
PI-SAC agents learn predictive information to improve RL efficiency.
problem Improving sample efficiency in reinforcement learning.
method PI-SAC agents use a contrastive version of Conditional Entropy Bottleneck to learn predictive information from past and future states.
result PI-SAC agents significantly improve sample efficiency on challenging continuous control tasks.
Deep reinforcement learning (RL) algorithms can use high-capacity deep networks to learn directly from image observations. However, these high-dimensional observation spaces present a number of challenges in practice, since the policy must now solve two problems: representation learning and task learning. In this work,…
Study tackles balancing policy switching costs in offline RL.
problem Balancing the cost of policy switching in offline RL.
method Optimal transport ideas and Net Actor-Critic algorithm.
result Demonstrated efficiency on multiple RL benchmarks.
Improved exploration in RL with latent state marginalization.
problem Complexity of deep probabilistic models limits their practical use in reinforcement learning.
method Adopting latent variable policies within the MaxEnt framework, with low-cost marginalization of latent states.
result Effective marginalization leads to better exploration and more robust training.
Paper tackles offline meta-reinforcement learning with a new algorithm.
problem Performing reinforcement learning on limited data from a new task.
method Meta-Actor Critic with Advantage Weighting (MACAW) algorithm.
result Achieves notable gains over prior methods on offline meta-RL benchmarks.
Develops RL for dynamic risk assessment in stochastic optimization.
problem Time-consistent risk assessment in stochastic optimization problems.
method Model-free reinforcement learning with dynamic convex risk measures, time-consistent dynamic programming, policy gradient updates, actor-critic neural network optimization.
result Demonstrates optimal policies for statistical arbitrage, financial hedging, and robot control.
PBVFs generalize across policies using learned value functions.
problem RL algorithms forget information about old policies when updating value functions to track the learned policy.
method Introduce Parameter-Based Value Functions (PBVFs) that include policy parameters in their inputs, enabling them to generalize across different policies.
result PBVFs enable zero-shot learning of new policies that outperform any policy seen during training.
Improved RL for grasping in cluttered scenes using state representation learning.
problem Poor performance of RL methods in grasping diverse objects from raw images.
method Employed state representation learning (SRL) with disentanglement of raw input images.
result Deep RL can learn grasping skills from varied visual inputs.
DyNODE uses neural ODEs to model system dynamics in continuous control tasks.
problem Modeling the dynamics of systems in continuous control tasks.
method Neural Ordinary Differential Equations (ODEs) combined with actor-critic RL.
result DyNODE outperforms standard neural networks in sample efficiency and predictive performance.
Designing optimal controllers continues to be challenging as systems are becoming complex and are inherently nonlinear. The principal advantage of reinforcement learning (RL) is its ability to learn from the interaction with the environment and provide optimal control strategy. In this paper, RL is explored in the cont…
RL teaches agents to make decisions through trial and error.
problem Sequential decision-making problems in dynamic environments.
method Value-based and policy search approaches, including actor-critic methods.
result RL can learn good policies from numeric reward feedback.
Self-supervised reward prediction improves RL in sparse reward settings.
problem Data efficiency and sparse reward signals in reinforcement learning.
method Learning a state representation for reward prediction and using it to shape rewards.
result Self-supervised reward prediction enhances RL algorithms in single-goal environments.
New RL method designs 3D molecules with improved symmetry.
problem Lack of 3D information in molecular design.
method Symmetry-aware actor-critic architecture using spherical harmonics.
result Improves generalization and molecule quality.
Cumulative entropy regularization introduces a regulatory signal to the reinforcement learning (RL) problem that encourages policies with high-entropy actions, which is equivalent to enforcing small deviations from a uniform reference marginal policy. This has been shown to improve exploration and robustness, and it ta…
MODULE solves LfO problem with high sample efficiency and stability.
problem Sample inefficiency and instability in LfO.
method Modified DSAC with MODULE algorithm.
result Superior performance in MuJoCo environments.
Pretraining with expert demonstrations have been found useful in speeding up the training process of deep reinforcement learning algorithms since less online simulation data is required. Some people use supervised learning to speed up the process of feature learning, others pretrain the policies by imitating expert dem…
This work tackles risk-sensitive deep RL by optimizing policies with variance constraints.
problem Risk and aleatoric uncertainty in deep reinforcement learning.
method Lagrangian and Fenchel dualities to transform the problem into an unconstrained saddle-point policy optimization problem, and an actor-critic algorithm to iteratively update policy, Lagrange multiplier, and Fenchel dual variable.
result The proposed actor-critic algorithm finds a globally optimal policy at a sublinear rate.
We introduce SLM Lab, a software framework for reproducible reinforcement learning (RL) research. SLM Lab implements a number of popular RL algorithms, provides synchronous and asynchronous parallel experiment execution, hyperparameter search, and result analysis. RL algorithms in SLM Lab are implemented in a modular w…
A new method for risk-sensitive reinforcement learning using Spectral Risk Measures.
problem Incorporating risk sensitivity into reinforcement learning algorithms.
method Proposes a novel framework for optimizing Spectral Risk Measures in both online and offline RL algorithms.
result Demonstrates consistent outperformance over existing risk-sensitive methods in various domains.
This paper proposes a new algorithm for learning guidance rewards in RL.
problem Long-term temporal credit assignment in sparse or delayed reward environments.
method Surrogate RL objective with trajectory-space smoothing to learn guidance rewards.
result Guidance rewards can be learned without additional neural networks and have intuitive interpretation.
Paper uses RL to optimize multi-asset portfolios in fluctuating markets.
problem Optimizing multi-asset portfolios in time-varying financial markets.
method Soft Actor-Critic (SAC) algorithm for policy learning, policy iteration process.
result SAC algorithm outperforms in various criteria in simulated and real financial markets.
New methods tackle robust reinforcement learning in sparse, corrupted data.
problem Tackles robust reinforcement learning in sparse, corrupted data.
method Proposes actor-critic methods with sparse robust estimator oracles.
result First non-vacuous guarantees in high-dimensional sparse MDPs with single-policy concentrability coverage.
We propose a general formulation for addressing reinforcement learning (RL) problems in settings with observational data. That is, we consider the problem of learning good policies solely from historical data in which unobserved factors (confounders) affect both observed actions and rewards. Our formulation allows us t…
Optimization of parameterized policies for reinforcement learning (RL) is an important and challenging problem in artificial intelligence. Among the most common approaches are algorithms based on gradient ascent of a score function representing discounted return. In this paper, we examine the role of these policy gradi…
ENIAC method optimizes and explores complex RL problems with non-linear policies.
problem Theoretical understanding of non-linear policies in RL with strategic exploration.
method ENIAC, an actor-critic method for non-linear function approximation.
result ENIAC finds near-optimal policies in polynomial exploration rounds under bounded eluder dimension.
Develops RL for optimal market-making in non-Markov processes.
problem Optimal market-making in non-Markov price processes.
method Deep reinforcement learning with Soft Actor-Critic (SAC) algorithm.
result Optimal strategy for market-making in semi-Markov and Hawkes Jump-Diffusion dynamics.
We establish a new connection between value and policy based reinforcement learning (RL) based on a relationship between softmax temporal value consistency and policy optimality under entropy regularization. Specifically, we show that softmax consistent action values correspond to optimal entropy regularized policy pro…
A new RL framework for risk-sensitive decision-making using convex scoring functions.
problem Time-inconsistent risk measures in reinforcement learning.
method Convex scoring functions, augmented state space, auxiliary variable, customized Actor-Critic algorithm.
result Theoretical guarantees for approximation and convergence under certain conditions.
Proposes a new framework for risk-sensitive RL using deep nets.
problem Risk-sensitive reinforcement learning problems.
method Conditional elicitability, scoring functions, deep neural networks.
result Dynamic spectral risk measures can be approximated by deep nets.
Model-free deep reinforcement learning (RL) algorithms have been successfully applied to a range of challenging sequential decision making and control tasks. However, these methods typically suffer from two major challenges: high sample complexity and brittleness to hyperparameters. Both of these challenges limit the a…
Markov Decision Processes (MDPs), the mathematical framework underlying most algorithms in Reinforcement Learning (RL), are often used in a way that wrongfully assumes that the state of an agent's environment does not change during action selection. As RL systems based on MDPs begin to find application in real-world sa…
We present an approach for reconfiguration of dynamic visual sensor networks with deep reinforcement learning (RL). Our RL agent uses a modified asynchronous advantage actor-critic framework and the recently proposed Relational Network module at the foundation of its network architecture. To address the issue of sample…
Paper uses SAC RL to optimize market-making strategies.
problem Optimizing market-making strategies with risk management.
method Applying SAC reinforcement learning to automate market-making decisions.
result Agent learns to optimize spreads and hedge trades.
ESAC improves reinforcement learning by lookahead and intuition.
problem Designing efficient reinforcement learning architectures for intelligent agents.
method Integrates selector, tuner, model-learner, estimator, and lookahead into ESAC.
result ESAC outperforms other architectures in optimizing policies.