This paper balances short-term and long-term rewards in policy learning.
problem Balancing short-term and long-term rewards in policy learning.
method Formalizes a new framework to balance rewards, identifies rewards under mild assumptions, deduces efficiency bounds, and develops a policy learning approach.
result The proposed method improves the estimator of long-term reward and reduces regret.
New algorithm optimizes for long-term user satisfaction in delayed reward settings.
problem Optimizing for long-term user satisfaction in delayed reward settings.
method Developed a predictive model of delayed rewards and a bandit algorithm that combines rewards and surrogate outcomes.
result Our algorithm significantly outperforms methods that optimize for short-term proxies or rely solely on delayed rewards.
Paper optimizes recommendation systems for long-term business metrics.
problem Short-term reward optimization ignores long-term business metrics.
method Introduced a framework for modeling long-term rewards in RecoGym.
result Proposed a simple extension leading to state-of-the-art results.
PoDiRe learns long-term rewards in multi-task recommendations.
problem Long-term rewards in multiple recommendation tasks.
method Policy Distilled Reinforcement Learning (PoDiRe) combining deep reinforcement learning and knowledge distillation.
result PoDiRe outperforms state-of-the-art methods in real-world data.
New framework estimates long-term outcomes from short-term data.
problem Estimating long-term outcomes from short-term data.
method Reward function decomposition-based framework (LOPE).
result LOPE outperforms existing methods, especially when surrogacy is violated.
Kernel method estimates long-term effects from short-term data.
problem Estimating long-term effects from short-term data in continuous actions.
method Kernel ridge regression to embed and extrapolate long-term effects.
result Uniform consistency and nonasymptotic error bounds for the estimator.
New algorithm optimizes long-term user satisfaction in recommendation systems.
problem Optimizing long-term user satisfaction in recommendation systems with delayed rewards.
method Developed a predictive model of delayed rewards and a bandit algorithm that balances exploration and exploitation.
result Our approach results in substantially better performance compared to short-term or delayed optimization.
A new approach for deep exploration in sparse reward reinforcement learning.
problem Slow or no learning in reinforcement learning with rare rewards.
method Long-term visitation count planning and decoupling exploration and exploitation.
result Significantly outperforms existing methods in sparse reward environments.
Finding optimal policies which maximize long term rewards of Markov Decision Processes requires the use of dynamic programming and backward induction to solve the Bellman optimality equation. However, many real-world problems require optimization of an objective that is non-linear in cumulative rewards for which dynami…
COCOA improves credit assignment in reinforcement learning by measuring contributions to rewards.
problem Improving sample efficiency in reinforcement learning through better credit assignment methods.
method Counterfactual Contribution Analysis (COCOA) for precise credit assignment.
result COCOA achieves lower bias and variance compared to Hindsight Credit Assignment (HCA), improving reinforcement learning performance.
New algorithm tackles nonstationary linear bandits with latent dynamics.
problem Nonstationary bandit problem with latent states and unknown dynamics.
method Explore-then-commit algorithm with exploration and commitment phases.
result Achieves ildeO(T2/3) regret. Study improves reinforcement learning for stable long-term performance.
problem Distributionally robust average-reward reinforcement learning for stable long-term performance.
method Proposes two algorithms to achieve near-optimal sample complexity.
result Achieves a sample complexity of O(∣S∣∣A∣tmix2ε−2) for estimating optimal policy and robust average reward. Paper proposes a model-free algorithm for CMDPs with long-term constraints, achieving optimal regret bounds.
problem Optimizing systems with long-term constraints where transition probabilities are unknown.
method Combines concepts from constrained optimization and Q-learning to propose an algorithm.
result Achieves optimal regret bounds for reward and constraint violation.
New synaptic model derived from reinforcement learning for spiking neurons.
problem Learning optimal actions in complex systems.
method Derives a synaptic update rule from reinforcement learning algorithms.
result Synaptic strengths lead to locally optimal reward values.
ARL uses queries to learn rewards, focusing on cost vs. reward value.
problem How to efficiently use queries to learn rewards in reinforcement learning.
method Proposed and evaluated heuristic approaches for ARL in multi-armed bandits and MDPs.
result Challenging aspects of ARL highlighted, including intractability of value computation.
This paper explores a simple regularizer for reinforcement learning by proposing Generative Adversarial Self-Imitation Learning (GASIL), which encourages the agent to imitate past good trajectories via generative adversarial imitation learning framework. Instead of directly maximizing rewards, GASIL focuses on reproduc…
Reward tweaking optimizes behavior for long-term goals by adjusting the reward function.
problem Optimizing behavior for long-term goals in reinforcement learning with unstable long planning horizons.
method Reward tweaking learns a surrogate reward function that induces optimal behavior for the original task.
result Reward tweaking guides agents towards better long-term returns while planning for short horizons.
Open-domain dialog generation is a challenging problem; maximum likelihood training can lead to repetitive outputs, models have difficulty tracking long-term conversational goals, and training on standard movie or online datasets may lead to the generation of inappropriate, biased, or offensive text. Reinforcement Lear…
There exist a number of reinforcement learning algorithms which learnby climbing the gradient of expected reward. Their long-runconvergence has been proved, even in partially observableenvironments with non-deterministic actions, and without the need fora system model. However, the variance of the gradient estimator ha…
New RL algorithms improve average-reward performance.
problem Improving RL algorithms for average-reward criteria.
method Developed novel algorithms addressing average-reward criterion directly.
result ATRPO significantly outperforms TRPO in challenging MuJuCo environments.
Paper uses RL to estimate long-term treatment effects efficiently.
problem Estimating long-term treatment effects in nonstationary settings.
method Reinforcement Learning (RL) for Markov processes.
result Demonstrated promising results in synthetic and real-world datasets.
The paper tackles batch policy learning in Markov Decision Processes, focusing on average reward maximization.
problem Maximizing long-term average reward in Markov Decision Processes with batch learning.
method Doubly robust estimator for average reward, optimization algorithm for optimal policy, finite-sample regret guarantee.
result The proposed method achieves semiparametric efficiency and provides a finite-sample regret guarantee.
Model user preferences for conversational LLMs using weak rewards.
problem Lack of persistent user models in conversational LLMs leading to repeated user restatements.
method Vector-Adapted Retrieval Scoring (VARS) framework that updates user vectors online from weak scalar rewards.
result Full VARS agent achieves strongest overall performance, matches strong Reflection baseline in task success, and reduces user effort.
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.
Reinforcement learning aids decision-making in economics and finance.
problem Optimal decision-making in dynamic, uncertain environments.
method Reinforcement learning algorithms to learn optimal policies.
result Deep learning enhances solving complex behavioral problems.
Algorithm improves movie recommendation efficiency with fairness constraints.
problem Improving movie recommendation efficiency with fairness constraints in combinatorial semi-bandits.
method Adopted Thompson Sampling with beta priors and Bernoulli likelihoods to handle fairness constraints.
result Time-averaged regret upper bounded by $\frac{N}{2η} + O\left(\frac{\sqrt{mNT\ln T}}{T}
ight)$, with fairness constraints satisfied.
Drawing an inspiration from behavioral studies of human decision making, we propose here a general parametric framework for a reinforcement learning problem, which extends the standard Q-learning approach to incorporate a two-stream framework of reward processing with biases biologically associated with several neurolo…
Algorithm reduces long-term policy regret in ML decision-making.
problem Capturing long-term impacts of ML decisions in communities.
method Modeling communities as arms in a multi-armed bandit problem, defining policy regret as a stronger metric than external regret.
result Algorithm achieves provably sub-linear policy regret for long time horizons.
Deep reinforcement learning has obtained significant breakthroughs in recent years. Most methods in deep-RL achieve good results via the maximization of the reward signal provided by the environment, typically in the form of discounted cumulative returns. Such reward signals represent the immediate feedback of a partic…
In model-based reinforcement learning, the agent interleaves between model learning and planning. These two components are inextricably intertwined. If the model is not able to provide sensible long-term prediction, the executed planner would exploit model flaws, which can yield catastrophic failures. This paper focuse…
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.
In this paper, we study reinforcement learning (RL) algorithms to solve real-world decision problems with the objective of maximizing the long-term reward as well as satisfying cumulative constraints. We propose a novel first-order policy optimization method, Interior-point Policy Optimization (IPO), which augments the…
Transformer learns long-term dependencies from real-world data.
problem Sample inefficiency in deep reinforcement learning.
method Transformer architecture applied to autoregressive real-world episodes.
result Transformer-based world model generates meaningful experience.
SDPG algorithm improves sample efficiency and reward in DRL for continuous action spaces.
problem Improving sample efficiency and reward in distributional reinforcement learning for continuous action spaces.
method SDPG algorithm models return distribution using samples via reparameterization technique.
result SDPG shows better sample efficiency and higher reward in OpenAI Gym environments.
We study a reinforcement learning setting, where the state transition function is a convex combination of a stochastic continuous function and a deterministic function. Such a setting generalizes the widely-studied stochastic state transition setting, namely the setting of deterministic policy gradient (DPG). We firstl…
There are great interests as well as many challenges in applying reinforcement learning (RL) to recommendation systems. In this setting, an online user is the environment; neither the reward function nor the environment dynamics are clearly defined, making the application of RL challenging. In this paper, we propose a …
Paper proposes a new risk measure (reward volatility) for optimizing financial decisions.
problem Managing uncertainty and volatility in financial decision-making.
method Defines reward volatility, derives policy gradient theorem, develops actor-only algorithm.
result Risk-averse optimization improves both reward volatility and return variance.
We extend Bayesian multi-armed bandit (MAB) algorithms beyond their original setting by making use of sequential Monte Carlo (SMC) methods. A MAB is a sequential decision making problem where the goal is to learn a policy that maximizes long term payoff, where only the reward of the executed action is observed. In the …
This work improves Q-learning for average-reward MDPs, reducing sample and communication complexities in federated settings.
problem Improving sample complexity of Q-learning for average-reward MDPs.
method Simple Q-learning algorithm with carefully chosen parameters for both single-agent and federated scenarios.
result Established first federated Q-learning algorithm for average-reward MDPs with provable efficiency in sample and communication complexities.
Reinforcement learning agents need exploratory behaviors to escape from local optima. These behaviors may include both immediate dithering perturbation and temporally consistent exploration. To achieve these, a stochastic policy model that is inherently consistent through a period of time is in desire, especially for t…
New algorithms for efficient learning with long-term rewards in contextual bandits.
problem Efficient learning with long-term rewards in contextual bandits.
method Proposes new algorithms leveraging sparsity to discover dependence patterns and arm parameters.
result Regret upper bounds for data-poor and data-rich regimes, showing improved sample complexity.
Constrained Markov Decision Process (CMDP) is a natural framework for reinforcement learning tasks with safety constraints, where agents learn a policy that maximizes the long-term reward while satisfying the constraints on the long-term cost. A canonical approach for solving CMDPs is the primal-dual method which updat…
New AI learns like neurons, generalizing from sparse rewards.
problem Designing AI that learns without explicit instructions and applies that learning to sparse reward scenarios.
method Combining neuroscience principles with computational efficiency, creating the Neurons-in-a-Box architecture.
result The architecture can learn efficiently and generalize across various tasks, including challenging environments.
A new algorithm tackles delayed combinatorial semi-bandit with causal relations.
problem Optimizing decisions in a non-stationary environment with delayed and causally related rewards.
method Formalized as a non-stationary delayed combinatorial semi-bandit problem, the approach models causal relations with a directed graph in a stationary structural equation model. The agent learns these relations from delayed feedback to optimize decisions.
result Proved a regret bound for the proposed algorithm's performance.
We propose and study a new model for reinforcement learning with rich observations, generalizing contextual bandits to sequential decision making. These models require an agent to take actions based on observations (features) with the goal of achieving long-term performance competitive with a large set of policies. To …
The paper addresses statistical inference for online decision-making in a contextual bandit setting.
problem Understanding the performance of reward models in online decision-making with contextual information.
method The paper uses the contextual bandit framework with a linear reward model and the ε-greedy policy to address the exploration-exploitation dilemma. It employs the martingale central limit theorem and inverse propensity score weighting to establish asymptotic normality of parameter estimators. result The online ordinary least squares estimator and the online weighted least squares estimator are asymptotically normal, providing insights into the performance of the reward model.
Develops first-order methods for average-reward MDPs with strong guarantees.
problem Lack of strong theoretical guarantees for first-order methods in AMDPs.
method Average-reward stochastic policy mirror descent (SPMD) and variance-reduced temporal difference (VRTD) methods.
result Establishes sample complexity results for solving AMDPs.
Reinforcement learning (RL) methods learn optimal decisions in the presence of a stationary environment. However, the stationary assumption on the environment is very restrictive. In many real world problems like traffic signal control, robotic applications, one often encounters situations with non-stationary environme…