Paper shows MCTS approximates policy optimization, proposing an improved variant.
problem Improving AI performance through better MCTS algorithms.
method Shows MCTS approximates policy optimization problem, proposes a new algorithm.
result Proposed algorithm reliably outperforms original AlphaZero in multiple domains.
Gradient-based methods are often used for policy optimization in deep reinforcement learning, despite being vulnerable to local optima and saddle points. Although gradient-free methods (e.g., genetic algorithms or evolution strategies) help mitigate these issues, poor initialization and local optima are still concerns …
Method learns optimal treatment policies from observational data.
problem Learning interpretable treatment assignment policies from observational data.
method Mixed-integer optimization (MIO) technology.
result Asymptotically exact in converging to optimal treatment policies.
AF improves classification models by adaptively weighting trees.
problem Improving classification model performance.
method AF combines OP2T for input-dependent weights and MIO for dynamic refinement.
result AF consistently outperforms RF, XGBoost, and other weighted RF.
PS framework selects best policy from library for CSO problems.
problem Policy selection in CSO with heterogeneous performance across covariate space.
method PS framework constructs library of candidate policies and learns a meta-policy to select the best one.
result PS consistently outperforms best single policy in heterogeneous CSO problems.
We propose a hybrid algorithmic strategy for complex stochastic optimization problems, which combines the use of scenario trees from multistage stochastic programming with machine learning techniques for learning a policy in the form of a statistical model, in the context of constrained vector-valued decisions. Such a …
Deep imagination optimizes decision-making in large trees with limited resources.
problem Optimal planning in large decision trees with limited resources and time.
method Analytical solutions and numerical analysis of sampling capacity allocation.
result Optimal policy is to allocate few samples per level for deep exploration, favoring depth over breadth.
While deep reinforcement learning has successfully solved many challenging control tasks, its real-world applicability has been limited by the inability to ensure the safety of learned policies. We propose an approach to verifiable reinforcement learning by training decision tree policies, which can represent complex p…
Boosting for off-policy learning reduces empirical risk.
problem Learning from logged bandit feedback without labeled data.
method A boosting algorithm optimizing policy's expected reward.
result Excess empirical risk decreases with each round of boosting.
Decision trees are ubiquitous in machine learning for their ease of use and interpretability. Yet, these models are not typically employed in reinforcement learning as they cannot be updated online via stochastic gradient descent. We overcome this limitation by allowing for a gradient update over the entire tree that i…
The paper sets bounds on how much regret is unavoidable in adaptive LQR with unknown B-matrix.
problem Understanding the limits of adaptive LQR with unknown B-matrix.
method Local asymptotic minimax regret lower bounds using van Trees' inequality and Bellman error representation.
result Logarithmic regret is impossible if the parametrization induces an uninformative optimal policy.
Finite-horizon lookahead policies are abundantly used in Reinforcement Learning and demonstrate impressive empirical success. Usually, the lookahead policies are implemented with specific planning methods such as Monte Carlo Tree Search (e.g. in AlphaZero). Referring to the planning problem as tree search, a reasonable…
In many settings, a decision-maker wishes to learn a rule, or policy, that maps from observable characteristics of an individual to an action. Examples include selecting offers, prices, advertisements, or emails to send to consumers, as well as the problem of determining which medication to prescribe to a patient. Whil…
CAPITAL algorithm identifies optimal patient subgroups for better treatment.
problem Identify maximum number of patients benefiting from better treatment.
method Constrained Policy Tree Search (CAPITAL) algorithm to find optimal subgroup selection rule (SSR).
result Maximizes the number of patients with enhanced treatment effects.
We address the problem of Bayesian reinforcement learning using efficient model-based online planning. We propose an optimism-free Bayes-adaptive algorithm to induce deeper and sparser exploration with a theoretical bound on its performance relative to the Bayes optimal policy, with a lower computational complexity. Th…
Monte-Carlo Tree Search (MCTS) methods are drawing great interest after yielding breakthrough results in computer Go. This paper proposes a Bayesian approach to MCTS that is inspired by distributionfree approaches such as UCT [13], yet significantly differs in important respects. The Bayesian framework allows potential…
This paper uses NLDT to find interpretable control rules from complex DRL policies.
problem Complex, non-interpretable policies from black-box AI methods.
method Evolutionary optimization of NLDT for hierarchical control rules.
result Interpretable control rules with similar performance to black-box DRL.
Monte Carlo Tree Search (MCTS) algorithms perform simulation-based search to improve policies online. During search, the simulation policy is adapted to explore the most promising lines of play. MCTS has been used by state-of-the-art programs for many problems, however a disadvantage to MCTS is that it estimates the va…
New method optimizes experiments under constraints.
problem Adapting BED to dynamic constraints in real-world tasks.
method Offline pre-training of an amortized policy and posterior network with online multi-step lookahead planning.
result Significantly more informative design sequences than existing methods.
New method learns better branching policies for MILP problems.
problem Improving branch and bound search for solving MILP problems.
method Imitates strong branching rule with parameterized state of B&B search tree.
result Generalized policies outperform current state-of-the-art.
Recently, a novel class of Approximate Policy Iteration (API) algorithms have demonstrated impressive practical performance (e.g., ExIt from [2], AlphaGo-Zero from [27]). This new family of algorithms maintains, and alternately optimizes, two policies: a fast, reactive policy (e.g., a deep neural network) deployed at t…
A RL approach finds Nash equilibrium for turn-based zero-sum games.
problem Finding Nash equilibrium in two-player turn-based zero-sum games.
method EIS method combining exploration, policy improvement, and supervised learning.
result EIS method finds an ε-approximate value function of Nash equilibrium in O(ε^(-(d+4))) steps.
This paper improves self-play learning in games by manipulating experience distributions.
problem Improving self-play learning in games through better experience sampling.
method Three approaches: weighted sampling, Prioritized Experience Replay, and diversifying trajectories.
result Major improvements in early training performance in some games, minor improvements overall.
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…
Monte Carlo Tree Search improves financial derivative hedging efficiency.
problem Optimizing pricing and hedging of derivative contracts in incomplete markets.
method Integrates tree search techniques with Reinforcement Learning for optimal control problems.
result Monte Carlo Tree Search outperforms Q-learning in sample efficiency and learning speed. InfoTree improves reinforcement learning by optimizing tool use with a greedy submodular approach.
problem Maximizing information from tool use in reinforcement learning with limited resources.
method Formalizes Rollout Informativeness, recasts state selection as submodular maximization, and uses UUCB and ABA.
result InfoTree outperforms existing methods across various benchmarks, improving performance by 18.2% on average.
The paper studies how neural policies can be interpreted using decision trees.
problem Understanding how machine learning controllers make decisions in complex environments.
method The approach involves disentangled representation using decision trees to interpret neural policies.
result The paper shows that disentanglement of learned neural dynamics improves interpretability.
Efficiently optimizes expensive functions with multi-step lookahead using one-shot optimization.
problem Optimizing expensive functions with long-term impacts using myopic approaches.
method Formulated as nested optimization problems within a multi-step scenario tree, optimized in one-shot fashion.
result Multi-step expected improvement is computationally tractable and outperforms existing methods.
The paper challenges the use of decision trees for pointwise inference due to slow convergence rates.
problem The slow convergence rates of decision trees in uniform norm, especially with non-vanishing probability.
method Demonstrates the limitations of adaptive recursive partitioning and shows how random forests can improve performance.
result Decision trees can fail to achieve polynomial rates of convergence in uniform norm, even with pruning.
BFTS uses Bayesian Additive Regression Trees for improved personalized mobile health interventions.
problem Adapting to complex, non-linear user behaviors in personalized mobile health interventions.
method Bayesian Forest Thompson Sampling (BFTS) integrates Bayesian Additive Regression Trees (BART) into the exploration loop of contextual bandits.
result BFTS achieves state-of-the-art regret on tabular benchmarks and improves engagement rates by over 30% in a behavioral intervention study.
Transformer learns to search through reinforcement learning, mimicking DFS.
problem Understanding how transformers learn search capabilities in RL.
method Two-head transformer, depth-wise curriculum, discounted returns.
result Transformer policy generalizes depth and prioritizes high-probability branches.
The study optimizes free trial lengths to boost subscriptions and consumer loyalty.
problem Optimizing free trial lengths to maximize customer acquisition and retention.
method A large-scale field experiment with personalized policy design and evaluation.
result Personalized free trial policies outperform uniform trial lengths.
Optimizes regret distribution in stochastic bandits for risk balance.
problem Balancing regret expectation and tail risk in stochastic bandits.
method Characterizes optimal regret tail probability for any threshold, proposes new policies.
result Discovers an intrinsic gap in optimal tail rate based on time horizon uncertainty.
New algorithm learns policies without uniform overlap assumption.
problem Learning optimal policies from non-uniformly collected data.
method Pessimistic Policy Learning (PPL) using lower confidence bounds.
result Efficient policy learning for adaptively collected data.
Improves RL planning by proposing sub-goals hierarchically.
problem Sequential planning assumption in RL.
method Divide-and-Conquer Monte Carlo Tree Search (DC-MCTS).
result Improves navigation and control tasks.
Bayesian model-based reinforcement learning is a formally elegant approach to learning optimal behaviour under model uncertainty, trading off exploration and exploitation in an ideal way. Unfortunately, finding the resulting Bayes-optimal policies is notoriously taxing, since the search space becomes enormous. In this …
Proposes a generalized causal tree for handling multiple treatments in uplift modeling.
problem Handling multiple treatments in uplift modeling.
method Generalizes causal tree algorithm to handle multiple discrete and continuous-valued treatments.
result Demonstrates improved performance over existing methods in experiments and real data examples.
In many areas, practitioners seek to use observational data to learn a treatment assignment policy that satisfies application-specific constraints, such as budget, fairness, simplicity, or other functional form constraints. For example, policies may be restricted to take the form of decision trees based on a limited se…
Reinforcement learning (RL) has recently been introduced to interactive recommender systems (IRS) because of its nature of learning from dynamic interactions and planning for long-run performance. As IRS is always with thousands of items to recommend (i.e., thousands of actions), most existing RL-based methods, however…
Paper tackles robust optimization under uncertainty using nested distance.
problem Optimizing under distributionally robust uncertainty with nested distance.
method Equivalent recursive and dynamic programming reformulations for tractable optimization.
result Optimal robust policies can be found efficiently using convex optimization.
Estimates sample size for subgroup analysis in randomized experiments.
problem Determining sample size for accurate subgroup analysis.
method Turns inference problem into simultaneous inference, calculates sample size based on confidence level and margin of error.
result Allows inversion of sample size to feasible number of treatment arms or partition complexity.
Simplifies complex pricing models for better interpretability and revenue.
problem Complex pricing models are hard to interpret and not widely adopted.
method Model distillation to create interpretable pricing policies.
result Maximizes revenue while maintaining interpretability.
Develops Heuristic Portfolio Optimization (HPO) as an information-restricted projection of Markowitz/tangency solution
problem Practitioners allocate capital with forecast-light rules like equal weight, inverse volatility, risk parity, HRP, and RA-HRP
method Implies-return principle and fixed-tree cluster-Sharpe recursion
result Formalizes HPO maps, proves defect equals squared inefficiency, and identifies nodewise alphas as policy-gradient coordinates
End-to-end policy learning method improves CATE estimation.
problem Learning optimal treatment policies from partially observed data.
method Modified causal forest for policy learning.
result Maximizing policy value is equivalent to minimizing CATE.
Paper uses CMAB to improve NAS efficiency and accuracy.
problem Improving efficiency and accuracy of NAS for DNNs.
method Formulated NAS as CMAB, used Nested Monte-Carlo Search.
result Discovered cell structure achieves comparable accuracy to state-of-the-art, 20x faster.
We demonstrate an application of risk-sensitive reinforcement learning to optimizing execution in limit order book markets. We represent taking order execution decisions based on limit order book knowledge by a Markov Decision Process; and train a trading agent in a market simulator, which emulates multi-agent interact…
Optimizes decision-making with uncertain variables using auxiliary observations.
problem Contextual stochastic optimization problems with uncertain variables and rich auxiliary observations.
method Trains forest decision policies by growing trees that optimize downstream decision quality, using optimization perturbation analysis for efficient approximations.
result Proves asymptotic optimality and empirical validation of the method's performance and efficiency.
Bayesian CART models improve insurance claims frequency prediction and interpretation.
problem Improving accuracy and interpretability in insurance pricing models.
method Introducing Bayesian CART models for claims frequency, implementing MCMC algorithm for posterior tree exploration, and using DIC for model selection.
result Bayesian CART models can better classify policy-holders into risk groups.