Optimal estimator derived for partially observable LTI systems.
problem Optimal estimator for partially observable LTI systems.
method State-space representation for derivation of optimal estimator.
result Derivation of minimum error variance estimator for partially observable LTI systems.
Bayesian approach optimizes in-context learning for state space models.
problem Optimizing in-context learning for state space models.
method Bayesian optimal sequential prediction over latent sequence tasks.
result Bayesian optimal predictor converges to posterior predictive mean.
Entropy regularization is used to get improved optimization performance in reinforcement learning tasks. A common form of regularization is to maximize policy entropy to avoid premature convergence and lead to more stochastic policies for exploration through action space. However, this does not ensure exploration in th…
New RL method handles large state-action spaces with complex models.
problem Complex models and large state-action spaces in reinforcement learning.
method π-KRVI, an optimistic modification of least-squares value iteration using kernel ridge regression.
result First order-optimal regret guarantees under general settings, improving over state of the art.
Paper introduces a new value function for state transitions and optimal policy learning.
problem Learning optimal policies from state transitions and actions.
method Develops a forward dynamics model to maximize a novel value function Q(s,s′). result Demonstrates benefits in value function transfer, redundant action spaces, and off-policy learning.
Optimal transport theory applied to quantum states on Grassmannians.
problem Developing optimal transport for quantum states.
method Metric geometry of Grassmannians and spectral theorem for density matrices.
result Wasserstein distance for normal states of von Neumann algebras.
This paper deals with discrete-time Markov control processes on a general state space. A long-run risk-sensitive average cost criterion is used as a performance measure. The one-step cost function is nonnegative and possibly unbounded. Using the vanishing discount factor approach, the optimality inequality and an optim…
Hill-ADAM optimizes loss landscapes by exploring state space deterministically.
problem Escaping local minima in loss landscapes.
method Hill-ADAM alternates between minimizing and maximizing error to explore the loss space.
result Hill-ADAM finds the global minimum state in loss landscapes.
We propose an algorithm for deterministic continuous Markov Decision Processes with sparse rewards that computes the optimal policy exactly with no dependency on the size of the state space. The algorithm has time complexity of O(∣R∣3×∣A∣2) and memory complexity of O(∣R∣×∣A∣), where ∣R∣ is the…
Generically learns movement control policies from exploration data.
problem Movement optimization in physically based characters.
method Parameterizes actions as target states, learns low-level control policy.
result Improves movement optimization across multiple tasks and algorithms.
Efficient RL in large POMDPs with latent determinism and embeddings.
problem Efficient reinforcement learning in large-scale POMDPs with latent states and observations.
method Conditional Hilbert space embeddings, linear optimal Q-function, deterministic latent transitions, gap assumption. result Computationally and statistically efficient algorithm for exact optimal policy.
We present a novel technique to solve the problem of managing optimally a pumped hydroelectric storage system. This technique relies on representing the system as a stochastic optimal control problem with state constraints, these latter corresponding to the finite volume of the reservoirs. Following the recent level-se…
Dynamical systems with large state-spaces are often expensive to thoroughly explore experimentally. Coarse-graining methods aim to define simpler systems which are more amenable to analysis and exploration; most current methods, however, focus on a priori state aggregation based on similarities in transition rates, whi…
Paper solves POMDPs in continuous time and discrete spaces.
problem Optimal decision making in discrete state and action space systems under partial observability.
method Combining optimal filtering theory and deep learning to solve a Hamilton-Jacobi-Bellman equation.
result Derives a mathematical description and solution approach for continuous-time POMDPs.
Unified framework extends adjoint Schrödinger bridge sampler to discrete spaces.
problem Challenges in learning discrete neural samplers due to gradients and combinatorial complexity.
method Introduces discrete ASBS, a unified framework that extends adjoint Schrödinger bridge sampler to discrete spaces.
result Empirically, discrete ASBS achieves competitive sample quality with significant advantages in training efficiency and scalability.
A DRL framework optimizes portfolios using a LFSS module for feature extraction.
problem Optimizing dynamic portfolios in financial markets.
method Deep Reinforcement Learning with a Latent Feature State Space module.
result The proposed DRL framework outperforms benchmarks in portfolio optimization.
AE-LSVI identifies near-optimal policies in complex systems with minimal data.
problem Identifying near-optimal policies in complex, costly data acquisition systems.
method Combines optimism and pessimism for active exploration in a generative model setting.
result Proves near-optimal policy identification over entire state spaces with polynomial sample complexity.
Sample inefficiency is a long-lasting problem in reinforcement learning (RL). The state-of-the-art estimates the optimal action values while it usually involves an extensive search over the state-action space and unstable optimization. Towards the sample-efficient RL, we propose ranking policy gradient (RPG), a policy …
The paper tackles reinforcement learning with exogenous variables and rewards.
problem Exogenous state variables and rewards slow reinforcement learning by introducing uncontrolled variation.
method Formalizes exogenous state variables and rewards, decomposes MDP into exogenous and endogenous components, and introduces algorithms to discover these components.
result Optimal policies for the endogenous MDP are also optimal for the original MDP, but the endogenous MDP is easier to solve due to reduced variance.
Operator calculus for population-based optimization provides a unified framework for analyzing convergence of various methods.
problem Convergence analysis of population-based optimization methods
method Introduce an operator calculus for describing composite mean-field algorithms as compositions of elementary operators acting on probability measures.
result Establish a modular Lyapunov principle for certifying exponential decay of state-space Lyapunov function and search errors.
Active learning selects inputs for GPSSM to learn latent states.
problem Optimally learn latent states of a GPSSM through active selection of inputs.
method Use mutual information to select informative inputs; approximate mutual information for GPSSM.
result Effective active learning of GPSSM dynamics in physical systems.
We study online reinforcement learning for finite-horizon deterministic control systems with {\it arbitrary} state and action spaces. Suppose that the transition dynamics and reward function is unknown, but the state and action space is endowed with a metric that characterizes the proximity between different states and…
We present a method for a certain class of Markov Decision Processes (MDPs) that can relate the optimal policy back to one or more reward sources in the environment. For a given initial state, without fully computing the value function, q-value function, or the optimal policy the algorithm can determine which rewards w…
Paper introduces multitask neural networks for efficient stochastic control problems.
problem Infeasibility of simulating state variables in some stochastic control problems.
method Multitask neural networks with dynamic task balancing.
result Multitask neural networks outperform state-of-the-art approaches in derivatives pricing problems.
Develops a regression approach for solving MDPs with general state and action spaces.
problem Solving MDPs with large or infinite state and action spaces.
method Regression-based primal-dual martingale approach.
result Tight upper and lower approximations of value functions and optimal policies.
New RL method reduces sample complexity for large state-action spaces.
problem Handling large state-action spaces in RL with general Q-functions.
method Nonparametric Q-learning using kernel ridge regression.
result Sample complexity is order optimal with respect to ε and kernel complexity.
New method learns state embeddings from demonstrations for improved reinforcement learning.
problem Difficult relationship between observed state and useful policy actions in dynamic problems.
method Variational framework for learning state embeddings that optimize trajectory linearity.
result Learning embedding spaces improves policy gradient reinforcement learning performance.
SMEs provide a transparent testbed for RL evaluation.
problem Lack of precise, white-box diagnostics in RL environments.
method Synthetic Monitoring Environments (SMEs) with fully configurable task characteristics and known optimal policies.
result SMEs allow for precise evaluation of RL algorithms, revealing the impact of specific environmental properties.
This work tackles large action spaces in RL by binarizing actions.
problem Large action spaces in reinforcement learning cause significant challenges.
method Sequentializing actions and binarizing the action space.
result Binarizing the action space can significantly improve RL algorithms and reduce state space size.
Predictability enables efficient parallelization of nonlinear models.
problem Understanding which nonlinear state space models can be efficiently parallelized.
method Established a relationship between system dynamics and optimization problem conditioning, quantified by the largest Lyapunov exponent.
result Predictable systems can be evaluated in O((logT)2) time, improving over conventional sequential approaches. A new method shapes reinforcement learning environments by abstracting large state spaces.
problem Learning in large, noisy environments with sparse feedback.
method Environment shaping using state abstraction.
result Agent's policy in shaped environment preserves near-optimal behavior in original environment.
Using stochastic gradient search and the optimal filter derivative, it is possible to perform recursive (i.e., online) maximum likelihood estimation in a non-linear state-space model. As the optimal filter and its derivative are analytically intractable for such a model, they need to be approximated numerically. In [Po…
A nonparametric approach for policy learning for POMDPs is proposed. The approach represents distributions over the states, observations, and actions as embeddings in feature spaces, which are reproducing kernel Hilbert spaces. Distributions over states given the observations are obtained by applying the kernel Bayes' …
We present the first PAC optimal algorithm for Bayes-Adaptive Markov Decision Processes (BAMDPs) in continuous state and action spaces, to the best of our knowledge. The BAMDP framework elegantly addresses model uncertainty by incorporating Bayesian belief updates into long-term expected return. However, computing an e…
A method for robust reinforcement learning in large state spaces.
problem Challenges in RL with large state spaces, costly data, and real-world dynamics deviation.
method Distributionally robust Markov decision processes with Gaussian Processes and maximum variance reduction.
result Efficient learning of multi-output nominal transition dynamics with statistical sample complexity bounds.
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 present an efficient algorithm for model-free episodic reinforcement learning on large (potentially continuous) state-action spaces. Our algorithm is based on a novel Q-learning policy with adaptive data-driven discretization. The central idea is to maintain a finer partition of the state-action space in regions w…
Hybrid model improves sequential data prediction by combining neural and time series models.
problem Nonlinear prediction in online settings with domain-specific feature engineering issues.
method Joint optimization of LSTM for feature extraction and SARIMAX for time series data using state space representations.
result Significant improvements in real-life competition datasets.
Consider a Markov decision process (MDP) that admits a set of state-action features, which can linearly express the process's probabilistic transition model. We propose a parametric Q-learning algorithm that finds an approximate-optimal policy using a sample size proportional to the feature dimension K and invariant …
A new active learning method for Gaussian process models.
problem Efficiently exploring unbounded state spaces for accurate models.
method Maximizes mutual information with respect to a bounded region using model predictive control.
result Our method yields a better model within the region of interest than entropy-based methods.
New optimizer G-AdaGrad improves upon AdaGrad for non-convex machine learning problems.
problem Solving non-convex machine learning problems efficiently.
method Proposes a new optimizer G-AdaGrad and analyzes its convergence using state-space models.
result Empirical results show G-AdaGrad performs better than AdaGrad and Adam.
We develop a normative framework for hierarchical model-based policy optimization based on applying second-order methods in the space of all possible state-action paths. The resulting natural path gradient performs policy updates in a manner which is sensitive to the long-range correlational structure of the induced st…
New RL method reduces sample complexity for large policy spaces.
problem Large-scale RL with unknown optimal policies and state/action spaces.
method Introduces eluder dimension for policy space, proving near-optimal sample complexity.
result Near-optimal sample complexity upper bound that depends linearly on eluder dimension.
RPO uses past and future state-action info for better policy optimization.
problem Sample inefficiency in on-policy reinforcement learning methods.
method Reflective Policy Optimization (RPO) integrates past and future state-action info for policy improvement.
result RPO improves policy performance and contracts the solution space, leading to faster convergence.
Develops a learning model predictive controller for competitive racing.
problem Lack of exploration in state space and complexity in obstacle avoidance.
method Explores state space through multiple initializations and develops a new method for convex terminal set selection.
result Yields a richer terminal safe set and maintains convexity.
Most real-world problems have huge state and/or action spaces. Therefore, a naive application of existing tabular solution methods is not tractable on such problems. Nonetheless, these solution methods are quite useful if an agent has access to a relatively small state-action space homomorphism of the true environment …
The policy gradient theorem is defined based on an objective with respect to the initial distribution over states. In the discounted case, this results in policies that are optimal for one distribution over initial states, but may not be uniformly optimal for others, no matter where the agent starts from. Furthermore, …
State-space models are used in a wide range of time series analysis formulations. Kalman filtering and smoothing are work-horse algorithms in these settings. While classic algorithms assume Gaussian errors to simplify estimation, recent advances use a broader range of optimization formulations to allow outlier-robust e…