TPGR uses a tree structure to improve efficiency and effectiveness in large-scale interactive recommendation.
problem Efficiency and effectiveness in large-scale interactive recommendation systems with thousands of items.
method Tree-structured Policy Gradient (TPGR) framework for handling large discrete action spaces.
result Superior recommendation performance and significant efficiency improvement over state-of-the-art methods.
Policy Prediction Network improves continuous control problems with model-free and model-based learning.
problem Improving sample complexity and performance in continuous control problems.
method Integrates model-free and model-based reinforcement learning, introduces implicit model-based learning for continuous action space.
result First to introduce implicit model-based learning to Policy Gradient algorithms for continuous action space.
Additive models, such as produced by gradient boosting, and full interaction models, such as classification and regression trees (CART), are widely used algorithms that have been investigated largely in isolation. We show that these models exist along a spectrum, revealing never-before-known connections between these t…
This paper proposes an online tree-based Bayesian approach for reinforcement learning. For inference, we employ a generalised context tree model. This defines a distribution on multivariate Gaussian piecewise-linear models, which can be updated in closed form. The tree structure itself is constructed using the cover tr…
Neural networks can learn Boolean circuits with local correlation.
problem Learning Boolean circuits with neural networks is computationally hard.
method Observing local correlation between input patterns and target labels, focusing on tree-structured Boolean circuits.
result Local correlation determines the success or failure of optimization in learning Boolean circuits.
Policy gradient aims to maximize expected return using gradient ascent.
problem Finding a policy that maximizes expected return in a given class of policies.
method Gradient ascent applied to a differentiable model of the policy, estimating the gradient of expected return.
result Policy gradient methods require on-policy data for gradient estimation, limiting sample efficiency.
Off-policy stochastic actor-critic methods rely on approximating the stochastic policy gradient in order to derive an optimal policy. One may also derive the optimal policy by approximating the action-value gradient. The use of action-value gradients is desirable as policy improvement occurs along the direction of stee…
AlphaCFG discovers alpha factors using grammar-guided search.
problem Discovering formulaic alpha factors in finance.
method AlphaCFG uses a grammar-based framework to define and discover alpha factors with syntactic and semantic constraints.
result AlphaCFG outperforms state-of-the-art methods in trading profitability and efficiency.
ARM policy gradient reduces variance for binary actions.
problem High variance in policy gradients for binary actions.
method Augment-Reinforce-Merge (ARM) policy gradient estimator.
result ARM estimator achieves significant variance reduction and faster convergence.
A new theorem and algorithm solve off-policy policy gradient problems.
problem Solving the theoretical gap in off-policy policy gradient methods.
method Introduced an off-policy policy gradient theorem using emphatic weightings and developed the ACE algorithm.
result Demonstrated ACE finds the optimal solution in off-policy learning, unlike previous methods.
DE via conjugate policies improves exploration and policy performance.
problem Effective exploration in policy gradient methods.
method DE via conjugate policies.
result DE improves policy performance and exploration effectiveness.
The problem of learning tree-structured Gaussian graphical models from independent and identically distributed (i.i.d.) samples is considered. The influence of the tree structure and the parameters of the Gaussian distribution on the learning rate as the number of samples increases is discussed. Specifically, the error…
New tree structure for pseudo-Anosovs from interval maps.
problem Understanding pseudo-Anosovs from interval maps.
method Tree structure on pseudo-Anosovs using rational numbers.
result Deepened dictionary between invariants.
Study a risk model with tree-structured Poisson-Markov random field for rainfall events.
problem Dependence between rainfall frequencies in insurance portfolios.
method Tree-structured Markov random field with Poisson marginals.
result Asymptotic results for portfolio risk and risk allocation.
Faster policy learning via continuous-time gradients.
problem Efficiently estimating policy gradients for continuous-time systems.
method Approximating continuous-time gradients directly, using adaptive discretization.
result More efficient policy gradient estimator leads to faster learning.
STORM-PG uses momentum for faster policy gradient updates.
problem Improving policy gradient methods for reinforcement learning.
method Introduces STORM-PG, a SARAH-based algorithm with exponential moving average.
result Achieves O(1/ε3) sample complexity, matching best-known rate. Optimizes Thompson sampling policies using policy gradient methods.
problem Improving Thompson sampling in bandit problems.
method Applies policy gradient algorithms to optimize Thompson sampling policies.
result Direct policy search on Thompson sampling improves performance.
Paper tackles efficient policy gradient estimation from off-policy data.
problem Estimating policy gradients from off-policy data is challenging and inefficient.
method Derives asymptotic lower bounds, proposes a meta-algorithm with 3-way robustness, and establishes convergence guarantees.
result Meta-algorithm achieves the lower bound on mean-squared error without parametric assumptions.
Policy gradient methods achieve linear convergence in simple MDPs.
problem Analyzing convergence rates of policy gradient methods in finite MDPs.
method Connections with policy iteration to show linear convergence with large step-sizes.
result Policy gradient methods succeed with large step-sizes and achieve linear rate of convergence.
DBQPG improves policy gradient estimation with fewer samples.
problem Accurate policy gradient estimation with limited samples.
method Deep Bayesian Quadrature Policy Gradient (DBQPG).
result DBQPG provides more accurate and less variable gradient estimates.
Deep sum-product networks learn faster than shallow models.
problem The speed of parameter optimization in sum-product networks.
method Theoretical analysis and empirical experiments on overparameterized sum-product networks.
result Gradient-based optimization in deep sum-product networks is equivalent to gradient ascent with adaptive and time-varying learning rates and additional momentum terms.
New k-step policy gradient method avoids local optima in restricted policy classes.
problem Suboptimal local optima in policy gradient methods for restricted policy classes.
method Proposes a k-step policy gradient method to escape myopic local optima. result The method converges to near optimal solutions exponentially close to the optimal deterministic policy.
New method corrects state distribution mismatch for off-policy policy optimization.
problem Mismatch between behavior and evaluation policy state distributions.
method Off-policy policy gradient with state distribution correction.
result Significantly improved policy quality in simulations.
The paper interprets policy-gradient algorithms using continuation theory.
problem Optimizing nonconvex functions in reinforcement learning.
method Formulates policy optimization as optimization by continuation, interprets policy-gradient algorithms as implicitly optimizing deterministic policies.
result Exploration in policy-gradient algorithms is seen as computing a continuation of the return of the policy.
Large deviations theory applied to policy gradient methods.
problem Understanding convergence of policy gradient methods in reinforcement learning.
method Large deviation rate function and contraction principle from large deviations theory.
result Convergence properties of policy gradient methods can be extended to various policy parametrizations.
We propose expected policy gradients (EPG), which unify stochastic policy gradients (SPG) and deterministic policy gradients (DPG) for reinforcement learning. Inspired by expected sarsa, EPG integrates across the action when estimating the gradient, instead of relying only on the action in the sampled trajectory. We es…
Proposes CoPO, a new policy optimization method for competitive games.
problem Designing efficient optimization methods for competitive Markov decision processes.
method Competitive policy optimization (CoPO) approach that exploits game-theoretic nature of competitive games.
result Stable optimization, convergence to sophisticated strategies, and higher scores compared to baseline methods.
We propose expected policy gradients (EPG), which unify stochastic policy gradients (SPG) and deterministic policy gradients (DPG) for reinforcement learning. Inspired by expected sarsa, EPG integrates (or sums) across actions when estimating the gradient, instead of relying only on the action in the sampled trajectory…
Develops DPG methods for continuous-time RL with deterministic policies.
problem High variance and slow convergence in stochastic policy RL methods.
method Derives continuous-time policy gradient formula and proposes CT-DDPG algorithm.
result CT-DDPG achieves superior stability and faster convergence in continuous-time RL.
Recent advances in policy gradient methods and deep learning have demonstrated their applicability for complex reinforcement learning problems. However, the variance of the performance gradient estimates obtained from the simulation is often excessive, leading to poor sample efficiency. In this paper, we apply the stoc…
Unified reinforcement learning objective improves policy performance prediction.
problem Improving policy performance prediction in reinforcement learning.
method Proposed a counterfactual objective and an emphatic approach to compute unbiased policy gradients.
result Geoff-PAC algorithm demonstrates empirical success in deep RL benchmarks.
The policy gradient approach is a flexible and powerful reinforcement learning method particularly for problems with continuous actions such as robot control. A common challenge in this scenario is how to reduce the variance of policy gradient estimates for reliable policy updates. In this paper, we combine the followi…
DSPI connects natural policy gradient to policy iteration, proving global convergence.
problem Optimizing policies in reinforcement learning.
method DSPI framework, combining smoothed policy iteration and natural policy gradient.
result DSPI achieves geometric convergence and optimal complexity for policy optimization.
Paper proposes a policy gradient method for confounded POMDPs.
problem Estimating policy gradients for confounded POMDPs with continuous state and observation spaces.
method Developed a novel identification result to estimate policy gradients using offline data, solved conditional moment restrictions, and applied min-max learning with function approximation.
result Showed global convergence of the proposed algorithm in finding the optimal policy.
Paper learns tree-structured models from distributed data efficiently.
problem Learning tree-structured Gaussian graphical models from distributed data with limited communication.
method Communication-efficient strategies for central node to learn structure from distributed data.
result Tree structure can be reliably recovered even with limited information (signs of local data samples).
End-to-end learning framework for tree-structured data.
problem Learning models struggle with tree-structured data due to lack of fixed-length vectors.
method Developed a novel framework for generic semantic tree-structured data of arbitrary topology.
result Framework yields comparable performance to standard models with dedicated feature-vectors and outperforms in compositional data.
Proposes log density gradient to improve reinforcement learning sample complexity.
problem Residual error in gradient estimation in policy gradient methods.
method Log density gradient method to correct residual error, using state-action discounted distributional formulation.
result Min-max optimization method to approximate log density gradient with on-policy samples, achieving sample complexity of m−1/2. New methods estimate policy value and gradients for deterministic policies from off-policy data.
problem Estimating policy value and gradients for deterministic policies from off-policy data.
method Proposed new doubly robust estimators based on kernelization approaches.
result Demonstrated a rate independent of horizon length for policy value and gradient estimation.
Policy gradient methods with aggregated states can achieve better performance than approximate policy iteration.
problem Approximation errors in policy and value function approximations.
method State-aggregated representations and policy gradient methods.
result Policy gradient methods can achieve a per-period regret bounded by ε, while approximate policy iteration and value iteration have a higher regret.
Neural policy gradient methods converge globally and sublinearly.
problem Global optimality and convergence of neural policy gradient methods.
method Actor-critic schemes with neural networks, proving global optimality and sublinear convergence rates.
result Neural natural and vanilla policy gradient methods converge to globally optimal policies and stationary points.
Softmax policy gradient achieves global optimality in wide neural networks with entropy regularization.
problem Optimizing softmax policies with neural networks in the mean-field regime.
method Modeling neural networks as Wasserstein gradient flows and proving global optimality of fixed points.
result Global optimality of softmax policy gradient in wide single hidden layer neural networks with entropy regularization.
This paper extends policy gradient methods to partially observable environments.
problem Learning optimal policies in partially observable environments.
method Developed new tools including advantage function to generalize policy gradient algorithms and study their convergence in partially observable Markovian policies.
result Generalized theoretical guarantees of policy gradient algorithms to partially observable domains.
Paper presents a new method for learning hyperbolic representations using tree structures.
problem Learning faithful low-dimensional hyperbolic embeddings of data.
method Metric-first approach to learn tree structure, then embed into hyperbolic manifold.
result Novel fast algorithm TreeRep learns tree approximating original metric.
Paper analyzes convergence of dynamic policy gradient for MDPs, improving performance in finite-time problems.
problem Optimal policies in finite-time MDPs are not stationary and require epoch-specific training.
method Introduces dynamic policy gradient combining dynamic programming and policy gradient, analyzes convergence for softmax parametrisation.
result Dynamic policy gradient training exploits finite-time structure, leading to better convergence bounds.
Trust-region methods and natural gradients are equivalent in certain policy search scenarios.
problem Improving policy search methods in continuous control tasks.
method Introducing compatible policy search (COPOS) that uses natural parameterization and compatible value function approximation to control entropy loss.
result COPOS yields state-of-the-art results in challenging tasks and reduces entropy loss.
Policy gradient is an efficient technique for improving a policy in a reinforcement learning setting. However, vanilla online variants are on-policy only and not able to take advantage of off-policy data. In this paper we describe a new technique that combines policy gradient with off-policy Q-learning, drawing experie…
Optimal rates for learning hidden tree structures are determined.
problem Learning hidden tree structures from noisy data.
method Study of the (noisy) information threshold and the Chow-Liu algorithm.
result Optimal rates for structure recovery are inversely proportional to the information threshold squared.
MAGE optimizes policies using action gradients from model-based learning.
problem Lack of direct gradient information from critics in actor-critic methods.
method Model-based actor-critic algorithm that learns action-value gradient.
result MAGE outperforms model-free and model-based baselines on continuous control tasks.