New IRL algorithm for continuous state spaces with formal guarantees.
problem Finding a reward function for expert behavior in continuous state spaces.
method Modeling the system using orthonormal functions and providing correctness proofs.
result Proof of correctness and formal guarantees on sample and time complexity.
Novel framework proves fast RL convergence in continuous spaces.
problem Analyzing stability in continuous state-action RL.
method Introduces a novel framework to analyze stability properties of RL.
result Highlights two key stability properties and demonstrates their satisfaction in RL.
GPPSTD uses Gaussian Processes for efficient RL in continuous states.
problem Efficient reinforcement learning in continuous state spaces.
method Gaussian Process Posterior Sampling Reinforcement Learning (GPPSTD) algorithm.
result Combining demonstration and exploration improves reinforcement learning efficiency.
RANDPOL uses randomized networks for efficient reinforcement learning in continuous state and action MDPs.
problem Efficient reinforcement learning in environments with continuous state and action spaces.
method RANDPOL uses randomized function approximation to represent policy and value functions, providing finite time guarantees and improved numerical performance.
result RANDPOL achieves better numerical performance and provides finite time guarantees compared to deep neural network based algorithms.
Bootstrap policies improve regret in continuous state-action reinforcement learning.
problem Improving regret in reinforcement learning for continuous state and action spaces.
method Bootstrap-based policies for stochastic linear systems with quadratic cost functions.
result Bootstrap policies achieve a square root scaling of regret with respect to time.
Extends reinforcement learning to continuous state spaces with safety constraints.
problem Safety-critical reinforcement learning in continuous state spaces with unknown dynamics.
method Introduces a novel Budgeted Bellman Optimality operator and applies it to continuous state spaces.
result Validated on spoken dialogue and autonomous driving applications.
Complex contagion model explains financial fire sales through continuous asset prices.
problem Modeling financial fire sales with a continuum of asset prices.
method Developed a threshold model of continuous-state cascades using real values for asset prices.
result Discretization approach accurately replicates the distribution of defaulted banks and asset prices.
We seek to learn an effective policy for a Markov Decision Process (MDP) with continuous states via Q-Learning. Given a set of basis functions over state action pairs we search for a corresponding set of linear weights that minimizes the mean Bellman residual. Our algorithm uses a Kalman filter model to estimate those …
Study robust control for systems with continuous states using adversarial perturbations.
problem Fragile policies in Markov control models under internal or external perturbations.
method Distributionally robust stochastic control with adaptive adversarial perturbations.
result Optimal robust policies for continuous state systems with uniform learning guarantees.
The paper tackles finding optimal treatment sequences in continuous state spaces.
problem Finding counterfactually optimal action sequences in continuous state spaces.
method Formalizes the problem using finite horizon Markov decision processes and structural causal models. Develops a search method based on the A* algorithm.
result The method can find optimal action sequences in polynomial time under certain conditions.
We are interested in understanding stability (almost sure boundedness) of stochastic approximation algorithms (SAs) driven by a `controlled Markov' process. Analyzing this class of algorithms is important, since many reinforcement learning (RL) algorithms can be cast as SAs driven by a `controlled Markov' process. In t…
New Q-learning methods improve sample complexity for MDPs with continuous states.
problem Challenges in MDPs with continuous state spaces.
method Two nearest neighbor Q-learning methods with improved sample complexities.
result Minimax optimal sample complexities for offline and online settings.
We introduce a framework for model learning and planning in stochastic domains with continuous state and action spaces and non-Gaussian transition models. It is efficient because (1) local models are estimated only when the planner requires them; (2) the planner focuses on the most relevant states to the current planni…
New algorithms learn MDPs with continuous states and actions using Gaussian processes.
problem Online learning in unknown, episodic MDPs with continuous states and actions.
method Developed variants of UCRL and posterior sampling algorithms using Gaussian process priors.
result Sublinear regret bounds for learning MDPs with specific kernel structures.
We protect value function approximators in continuous state spaces from differential privacy attacks.
problem Protecting reward information from inverse reinforcement learning attacks in continuous state spaces.
method Adding functional noise to the value function iteratively to ensure differential privacy.
result Rigorous privacy guarantees and approximate optimality of the algorithm in discrete state spaces.
MAC improves reinforcement learning by estimating action values directly.
problem Discrete-action continuous-state reinforcement learning variance reduction.
method MAC uses the agent's action value representation to estimate policy gradient, reducing variance.
result MAC reduces policy gradient variance compared to traditional methods.
This study bridges discrete and continuous state spaces using the Ehrenfest process and diffusion models.
problem Understanding the relationship between discrete and continuous state spaces in stochastic processes.
method Investigates time-continuous Markov jump processes on discrete state spaces and their correspondence to state-continuous diffusion processes.
result The time-reversal of the Ehrenfest process converges to the time-reversed Ornstein-Uhlenbeck process, bridging discrete and continuous state spaces.
Continuous state spaces and stochastic, switching dynamics characterize a number of rich, realworld domains, such as robot navigation across varying terrain. We describe a reinforcementlearning algorithm for learning in these domains and prove for certain environments the algorithm is probably approximately correct wit…
Unified framework for continuous-state discrete flow matching models.
problem Discrete generative modeling with continuous probabilities.
method Introducing α-Flow, a family of CS-DFM models based on information geometry. result Optimal flow matching loss for α-flow minimizes generalized kinetic energy. NNQL learns Q-function from single sample path in continuous state space MDPs.
problem Learning optimal Q-function in MDPs with unknown transition kernel and continuous state space.
method Nearest Neighbor Q-Learning (NNQL) using nearest neighbor regression.
result NNQL achieves ε-accurate Q-function estimation with ildeO(L/(ε3(1−γ)7)) samples. Continual state learning model using generative replay for RL.
problem Efficiently learn and adapt state representations as the environment changes.
method Variational Auto-Encoders for state representation and Generative Replay for past knowledge.
result Automatic environment change detection and efficient state representation.
OpenAI Five defeated Dota 2 champions using deep reinforcement learning.
problem Complex, continuous state-action spaces in Dota 2.
method Deep reinforcement learning, distributed training, self-play.
result Superhuman performance in Dota 2 achieved through reinforcement learning.
This paper improves reinforcement learning policies in a scalable way.
problem Ensuring monotonic policy improvement in entropy-regularized RL.
method Derives an entropy-aware lower bound and proposes a novel RL algorithm.
result Demonstrates effectiveness in continuous-state tasks using a linear function approximator.
New RL algorithm for large state spaces with explicit exploration and exploitation phases.
problem Reinforcement learning in large or infinite state spaces.
method Model-based approach with explicit exploration and exploitation phases.
result Proves near-optimal policy with polynomial sample complexity under certain assumptions.
Entangled bisimulation improves policy learning from visual input.
problem Learning generalizeable policies from visual input in the presence of visual distractions.
method Proposes entangled bisimulation, a bisimulation metric for continuous state and action spaces.
result Entangled bisimulation improves policy learning on the Distracting Control Suite (DCS).
New method speeds up Gibbs sampling for large graphs.
problem Efficiently sampling from large graphical models.
method Poisson-minibatching Gibbs sampling.
result Theoretical convergence rate guarantees for Poisson-minibatching Gibbs.
Bayes-CPACE optimally explores continuous BAMDPs.
problem Model uncertainty in continuous state and action spaces.
method Covering state-belief-action space with samples, exploiting Lipschitz continuity.
result Near-optimal value function computed efficiently.
Efficient Q-learning for metric spaces without planning oracles.
problem Efficient reinforcement learning in continuous state-action spaces.
method Model-free Q-learning adapted for metric spaces without a planning oracle.
result Sample efficiency in continuous state-action spaces.
Paper uses DRL for automated power allocation in satellites.
problem Manual resource allocation is impractical for satellites with many power degrees of freedom.
method Continuous state and action spaces, Proximal Policy Optimization (PPO) algorithm.
result DRL shows promising results for minimum Unmet System Demand and power consumption.
Efficient algorithm for reinforcement learning in large state-action spaces with adaptive discretization.
problem Efficient reinforcement learning in large, potentially continuous state-action spaces.
method Adaptive Q-learning policy with data-driven adaptive discretization. result Demonstrates improved performance compared to existing methods, especially in adapting to the problem's structure.
Data-efficient reinforcement learning (RL) in continuous state-action spaces using very high-dimensional observations remains a key challenge in developing fully autonomous systems. We consider a particularly important instance of this challenge, the pixels-to-torques problem, where an RL agent learns a closed-loop con…
The paper explains how continuous language models can produce discrete, interpretable meanings.
problem Semantic collapse in continuous systems of large language models.
method Formalizing large language models as Continuous State Machines (CSMs) and analyzing the associated transfer operator.
result The leading eigenfunctions of the transfer operator induce a finite number of invariant meaning basins, explaining how continuous computation can produce discrete, interpretable semantics.
First we provide a simple set of sufficient conditions for the weak convergence of scaled affine processes with state space R+×Rd. We specialize our result to one-dimensional continuous state branching processes with immigration. As an application, we study the asymptotic behavior of least squares estimators…
Paper explores using LLMs for zero-shot reinforcement learning in continuous spaces.
problem Leveraging LLMs for continuous state spaces in reinforcement learning.
method Disentangled In-Context Learning (DICL) to handle multivariate data and control signal.
result DICL produces well-calibrated uncertainty estimates in reinforcement learning settings.
MO2 learns useful behaviours from past experience for new tasks.
problem Discovering useful behaviours from past experience and transferring them to new tasks.
method Model-Based Offline Options (MO2) framework supporting sample-efficient bottleneck option discovery over continuous state-action spaces.
result MO2 outperforms recent option learning methods on complex long-horizon continuous control tasks.
S-TRIGGER learns state representations for continual learning.
problem Efficiently compress and maintain past knowledge in changing environments.
method Generative Replay with self-triggered environment change detection.
result S-TRIGGER enables fast and high-performing Reinforcement Learning without catastrophic forgetting.
DeepMDP simplifies complex observations into continuous latent states.
problem Learning from high-dimensional observations in reinforcement learning.
method Trains a DeepMDP model that predicts rewards and next latent states.
result Optimization of DeepMDP objectives ensures quality of latent space and environment model.
Data-efficient learning in continuous state-action spaces using very high-dimensional observations remains a key challenge in developing fully autonomous systems. In this paper, we consider one instance of this challenge, the pixels to torques problem, where an agent must learn a closed-loop control policy from pixel i…
This paper proposes a method to safely adjust exploration in RL to satisfy constraints.
problem Unsafe exploration in reinforcement learning violates constraints on controlled object states.
method Automatic adjustment of exploration inputs and variance-covariance matrix for safety.
result The method guarantees satisfaction of joint chance constraints with specified probability.
New RL approach tackles non-linear MDPs without linear assumptions.
problem Sample efficiency in RL for complex, nonlinear MDPs with continuous states.
method Introduces EPW condition to relax linear structure requirements; provides sample-efficient RL algorithm.
result EPW condition allows solving MDPs without linear assumptions, including Atari games.
In this paper we extend temporal difference policy evaluation algorithms to performance criteria that include the variance of the cumulative reward. Such criteria are useful for risk management, and are important in domains such as finance and process control. We propose both TD(0) and LSTD(lambda) variants with linear…
ACSSM models irregular time series with continuous dynamics.
problem Modeling irregular time series data.
method ACSSM uses a multi-marginal Doob's h-transform and variational inference with stochastic optimal control.
result ACSSM outperforms in tasks like classification, regression, interpolation, and extrapolation.
We show the existence and orthogonality of wave operators naturally associated to a compatible Laplacian on a complete manifold with a corner of codimension 2. In fact, we prove asymptotic completeness i.e. that the image of these wave operators is equal to the space of absolutely continuous states of the compatible La…
New algorithms improve exploration in MDPs with theoretical guarantees.
problem Efficient exploration in undiscounted MDPs with continuous states.
method Exploration bonuses for SCAL and C-SCAL algorithms.
result Achieves sublinear regret with improved computational efficiency.
Study on variance of policy gradient in simple RL environments.
problem Understanding variance of policy gradient estimators in continuous RL.
method Analyzes REINFORCE estimator in linear-quadratic environments with Gaussian noise.
result Derives and validates bounds on estimator variance empirically.
New algorithms learn in complex decision-making problems with smooth transitions.
problem Learning in complex decision-making problems with smooth transitions.
method UCB and PSRL philosophies applied to episodic Markov decision processes with kernel approximation.
result Low regret learning achieved in continuous state and action spaces.
Study on Q-function estimation for continuous state-action MDPs, deriving rates and conditions.
problem Estimating Q-function in off-policy evaluation for continuous state-action Markov decision processes. method Reformulated as nonparametric instrumental variables (NPIV) problem, derived minimax lower bounds, proposed sieve two-stage least squares estimator.
result First minimax lower bounds for Q-function and its derivatives in sup-norm and L2-norm, same as classical nonparametric regression. FORK improves model-free reinforcement learning performance.
problem Improving model-free reinforcement learning performance.
method Introducing a new forward-looking Actor (FORK) for Actor-Critic algorithms.
result FORK significantly improves performance in various environments.