Algorithm tackles constrained reinforcement learning with concave-convex and knapsack constraints.
problem Constrained episodic reinforcement learning with concave rewards and convex constraints.
method Modular analysis with strong theoretical guarantees for concave-convex and knapsack settings.
result Significantly outperforms existing approaches in constrained episodic environments.
Study preference-based reinforcement learning in episodic kernel MDPs.
problem Learning from episodic human preferences in reinforcement learning.
method Developed preference-based value estimation and confidence sets for kernel-based MDPs.
result Proved high-probability regret bounds that converge to optimal policy value.
MEMEC improves sample efficiency in reinforcement learning.
problem Lack of sample efficiency in reinforcement learning.
method Proposes MEMEC, a Boltzmann policy with state-dependent temperature for more principled exploration.
result MEMEC outperforms other methods on classic RL environments and Atari games.
Paper proposes RRD to learn proxy rewards for sparse delayed rewards in episodic reinforcement learning.
problem Learning from sparse and delayed rewards in reinforcement learning.
method Randomized Return Decomposition (RRD) algorithm to redistribute rewards.
result Substantial improvement over baseline algorithms in experiments.
Paper proposes DAC-ML, a cognitive architecture that learns quickly from few episodes.
problem Sample inefficiency in AI learning action policies.
method Incorporates hippocampus-inspired sequential memory system into DAC theory of mind.
result DAC-ML rapidly converges to effective action policies maximizing reward.
Long horizon reinforcement learning is as hard as short horizon learning.
problem Understanding the difficulty of long horizon reinforcement learning problems.
method Introduced new concepts: ε-net for optimal policies and Online Trajectory Synthesis algorithm.
result Proved that sample complexity scales logarithmically with the planning horizon, refuting the conjecture.
Quantum UCB algorithm reduces reinforcement learning regret exponentially.
problem Episodic reinforcement learning with quantum state evolution.
method Upper Confidence Bound (UCB) quantum algorithm with quantum mean estimation.
result Exponential improvement in regret from $\Tilde{\mathcal{O}}(\sqrt{K})$ to $\Tilde{\mathcal{O}}(1)$ .
Improved reinforcement learning for episodes with varying action sets.
problem Reinforcement learning with context-dependent action sets.
method Extends MVP algorithm to handle adversarial and stochastic contexts.
result Established minimax regret bounds of O ( S A H 3 K log L ) O(\sqrt{SAH^3K\log L}) O ( S A H 3 K log L ) for adversarial contexts. This paper explores how environmental properties can simplify reinforcement learning in non-episodic settings.
problem Challenges in reinforcement learning with continuous interaction and sparse delayed rewards.
method Analysis of environment shaping and dynamism properties to simplify learning.
result Properties like environment shaping and dynamism can significantly ease learning in non-episodic, sparse reward settings.
We propose Episodic Backward Update (EBU) - a novel deep reinforcement learning algorithm with a direct value propagation. In contrast to the conventional use of the experience replay with uniform random sampling, our agent samples a whole episode and successively propagates the value of a state to its previous states.…
UCB-Advantage learns MDPs with O ( H 2 S A T ) O(\sqrt{H^2SAT}) O ( H 2 S A T ) regret.
problem Model-free reinforcement learning in finite-horizon MDPs.
method Reference-Advantage decomposition for low regret.
result Achieves i l d e O ( H 2 S A T ) ilde{O}(\sqrt{H^2SAT}) i l d e O ( H 2 S A T ) regret, matching best known bounds. Recent advances in deep reinforcement learning algorithms have shown great potential and success for solving many challenging real-world problems, including Go game and robotic applications. Usually, these algorithms need a carefully designed reward function to guide training in each time step. However, in real world, …
New algorithm reduces reinforcement learning complexity, approaching contextual bandits.
problem Episodic reinforcement learning's difficulty compared to contextual bandits.
method Proposes MVP algorithm with a new Bernstein-type bonus for episodic reinforcement learning.
result Achieves near-optimal regret bound of $O\left(\left(\sqrt{SAK} + S^2A
ight) \poly\log \left(SAHK
ight)
ight)$ , improving state-of-the-art results.
Efficiently stores and retrieves past states for faster learning in reinforcement learning.
problem Data inefficiency and memory limitations in reinforcement learning.
method Dynamic online k-means for state clustering and prioritization.
result Dynamic online k-means improves performance with smaller memory sizes.
Study shows reinforcement learning is possible with once-per-episode feedback.
problem Challenges of reinforcement learning with sparse feedback.
method Introduced a model where trajectory labels are generated by an unknown parametric model and developed an algorithm for sublinear regret.
result Achieved sublinear regret with a statistically and computationally efficient algorithm.
We present a new deep meta reinforcement learner, which we call Deep Episodic Value Iteration (DEVI). DEVI uses a deep neural network to learn a similarity metric for a non-parametric model-based reinforcement learning algorithm. Our model is trained end-to-end via back-propagation. Despite being trained using the mode…
Most provably-efficient learning algorithms introduce optimism about poorly-understood states and actions to encourage exploration. We study an alternative approach for efficient exploration, posterior sampling for reinforcement learning (PSRL). This algorithm proceeds in repeated episodes of known duration. At the sta…
State of the art deep reinforcement learning algorithms take many millions of interactions to attain human-level performance. Humans, on the other hand, can very quickly exploit highly rewarding nuances of an environment upon first discovery. In the brain, such rapid learning is thought to depend on the hippocampus and…
New algorithm reduces reinforcement learning regret to sqrt(d^3T).
problem Efficient reinforcement learning with generalized linear function approximation.
method Optimistic closure assumption for generalized linear functions.
result Proved regret bound of O(sqrt(d^3T)).
New algorithms reduce reinforcement learning regret in factored MDPs.
problem Optimizing reinforcement learning in non-episodic factored MDPs.
method Proposed two near-optimal and oracle-efficient algorithms for FMDPs.
result Oracle-efficient algorithms achieve near-optimal regret bounds of O ( D S A T ) O(DS\sqrt{AT}) O ( D S A T ) . StepMix algorithm ensures safe exploration in reinforcement learning with near-optimal performance.
problem Conservative exploration in reinforcement learning under episode-wise constraints.
method StepMix algorithm balances exploitation and exploration while ensuring episode-wise conservative constraint.
result StepMix achieves near-optimal regret order as in the constraint-free setting.
New algorithm for average reward learning with bounded hitting time assumption.
problem Minimizing regret in average reward reinforcement learning with bounded hitting time.
method Optimistic Q-learning with a novel L ‾ \overline{L} L operator for bounded hitting time. result Regret bound of i l d e O ( H 5 S A T ) ilde{O}(H^5 S\sqrt{AT}) i l d e O ( H 5 S A T ) for average reward learning. Logarithmic regret achieved in continuous-time linear-quadratic reinforcement learning.
problem Optimizing control actions in unknown continuous-time systems over a finite time horizon.
method Least-squares algorithm based on continuous-time observations and controls, with perturbation analysis and parameter estimation error analysis.
result Logarithmic regret bound of order O ( ( ln M ) ( ln ln M ) ) O((\ln M)(\ln\ln M)) O (( ln M ) ( ln ln M )) . New algorithm reduces regret and constraint violation in adversarial CMDP learning.
problem Online learning for episodic stochastically constrained Markov decision processes (CMDPs) with adversarial loss.
method Upper Confidence Primal-Dual Reinforcement Learning (UC-PDL) algorithm.
result Achieves O ~ ( L ∣ S ∣ ∣ A ∣ T ) \widetilde{\mathcal{O}}(L|\mathcal{S}|\sqrt{|\mathcal{A}|T}) O ( L ∣ S ∣ ∣ A ∣ T ) upper bounds of both regret and constraint violation. Many interesting real world domains involve reinforcement learning (RL) in partially observable environments. Efficient learning in such domains is important, but existing sample complexity bounds for partially observable RL are at least exponential in the episode length. We give, to our knowledge, the first partially …
Deep reinforcement learning methods attain super-human performance in a wide range of environments. Such methods are grossly inefficient, often taking orders of magnitudes more data than humans to achieve reasonable performance. We propose Neural Episodic Control: a deep reinforcement learning agent that is able to rap…
Unified framework for model- and value-optimistic reinforcement learning.
problem Designing and analyzing algorithms for episodic reinforcement learning.
method Lagrangian duality for a unified view of optimism.
result Unified algorithms with both efficient implementation and probabilistic analysis.
Paper proposes new γ γ γ -regret measure for non-episodic RL.
problem Measuring performance in non-episodic RL environments.
method Introduces γ γ γ -regret as a new performance measure and derives bounds. result Closed the gap between lower and upper bounds for γ γ γ -regret. OPSRL algorithm reduces regret with few samples in reinforcement learning.
problem High regret in reinforcement learning with limited data.
method Optimistic Posterior Sampling (OPSRL) with logarithmic sample complexity.
result Guaranteed high-probability regret bound of O ~ ( H 3 S A T ) \widetilde{\mathcal{O}}(\sqrt{H^3SAT}) O ( H 3 S A T ) . New framework for reinforcement learning with sporadic state observations.
problem Partial observability in reinforcement learning.
method Action-Triggered Sporadically Traceable Markov Decision Processes (ATST-MDPs).
result Optimistic algorithm achieving regret bound for episodic learning.
We initiate the study of multi-stage episodic reinforcement learning under adversarial corruptions in both the rewards and the transition probabilities of the underlying system extending recent results for the special case of stochastic bandits. We provide a framework which modifies the aggressive exploration enjoyed b…
SMRL uses score matching for efficient RL with exponential family models.
problem Efficient reinforcement learning with unknown transition models.
method Score matching for parameter estimation of exponential family models.
result Achieves online regret of O ( d H 3 T ) O(d\sqrt{H^3T}) O ( d H 3 T ) . KeRNS tackles non-stationary reinforcement learning in metric spaces.
problem Non-stationary reinforcement learning in metric spaces.
method KeRNS uses time-dependent kernels to model non-stationary Markov Decision Processes (MDPs).
result KeRNS achieves a regret bound that scales with the covering dimension and total variation of the MDP.
This is a brief technical note to clarify some of the issues with applying the application of the algorithm posterior sampling for reinforcement learning (PSRL) in environments without fixed episodes. In particular, this paper aims to: - Review some of results which have been proven for finite horizon MDPs (Osband et a…
UC-SSP algorithm tackles exploration in stochastic shortest path problems without loop-free assumption.
problem Exploration in goal-oriented reinforcement learning problems under stochastic shortest path formulation.
method UC-SSP algorithm with a novel stopping rule to interrupt and switch policies.
result Regret bound of O ~ ( D S A D K ) \displaystyle \widetilde{\mathcal{O}}( D S \sqrt{ A D K}) O ( D S A D K ) after K K K episodes. Safe reinforcement learning tackles safety constraints with linear approximations.
problem Ensuring safety in reinforcement learning without violating constraints.
method Modeling safety as a linear cost function, developing SLUCB-QVI and RSLUCB-QVI algorithms for MDPs with linear function approximation.
result Achieved a nearly optimal regret bound for safe reinforcement learning, matching state-of-the-art unsafe algorithms.
Extends Thompson sampling for RL with fewer episodes.
problem Limited episodes in RL settings.
method Batch Bayesian optimization over episodes to learn action bias terms.
result Significantly outperforms standard Thompson sampling.
We propose a new reinforcement learning algorithm for partially observable Markov decision processes (POMDP) based on spectral decomposition methods. While spectral methods have been previously employed for consistent learning of (passive) latent variable models such as hidden Markov models, POMDPs are more challenging…
PS4POMDPs algorithm simplifies online learning for episodic POMDPs with unknown models.
problem Learning in POMDPs is harder than in MDPs; online learning is especially challenging.
method Posterior Sampling-based reinforcement learning algorithm (PS4POMDPs)
result Bayesian regret scales as √number of episodes and is polynomial in other parameters.
This paper studies poisoning attacks in episodic RL and discovers their effectiveness depends on reward bounds.
problem Understanding security threats to RL algorithms through poisoning attacks.
method Examined two types of poisoning attacks: reward and action manipulation, in bounded and unbounded reward settings.
result The effectiveness of poisoning attacks depends on reward bounds, with different attack costs and success rates.
New algorithm for online learning in episodic MDPs with convex objectives.
problem Online episodic convex reinforcement learning.
method Online mirror descent algorithm with varying constraint sets and exploration bonus.
result Near-optimal regret bounds for online CURL without prior knowledge of transition function.
UDRL fails to converge in stochastic environments with episodic resets.
problem UDRL's convergence in stochastic environments with resets is questioned.
method UDRL is a supervised learning approach that does not use value functions.
result UDRL diverges in a simple stochastic environment with resets.
Proposes a method to avoid excessive exploration in reinforcement learning.
problem Avoiding excessive exploration in reinforcement learning to deploy it in practice.
method Designs a novel algorithm using UCB reinforcement learning policy with adaptive exploration constraints.
result Proves that the approach remains conservative while minimizing regret in tabular settings and validates on real-world tasks.
Study risk-sensitive reinforcement learning with Lipschitz dynamic risk measures, establishing regret bounds.
problem Risk-sensitive reinforcement learning in Markov decision processes.
method Two model-based algorithms for Lipschitz dynamic risk measures, focusing on regret bounds.
result Upper bounds demonstrate optimal dependencies on actions and episodes, reflecting risk sensitivity vs. sample complexity trade-off.
Paper optimizes learning MDPs with adversarial rewards using linear models.
problem Learning MDPs with adversarial rewards and changing reward functions.
method Proposes POWERS, an optimistic policy optimization algorithm.
result Achieves nearly optimal regret bound of O ( d H T ) O(dH\sqrt{T}) O ( d H T ) . Reinforcement learning (RL) algorithms have made huge progress in recent years by leveraging the power of deep neural networks (DNN). Despite the success, deep RL algorithms are known to be sample inefficient, often requiring many rounds of interaction with the environments to obtain satisfactory performance. Recently,…
New framework for RL with linear-convex models reduces performance gap.
problem Continuous-time episodic reinforcement learning with unknown coefficients and convex objectives.
method Probabilistic framework and phase-based learning algorithm for optimal exploration-exploitation trade-off.
result Sublinear regrets achieved, matching best possible results in literature.
Study gap-dependent regret bounds for risk-sensitive RL.
problem Risk-sensitive reinforcement learning with entropic risk measure.
method Propose cascaded gaps to adapt to problem structures, derive regret bounds.
result Exponential improvement over existing bounds in appropriate settings.