arXiv research

Paper tackles constrained bandit problems with a new learning framework.

POLAR optimizes treatment strategies in dynamic settings with statistical guarantees.

Proposes DRRO to mitigate over-optimization in RLHF from human feedback.

Pessimistic RL algorithm improves offline RL performance.

It is well known that quantile regression model minimizes the portfolio extreme risk, whenever the attention is placed on the estimation of the response variable left quantiles. We show that, by considering the entire conditional distribution of the dependent variable, it is possible to optimize different risk and perf…

A new algorithm for offline RL with trajectory-wise reward reduces bias and variance errors.

Paper tackles robust decision-making from multiple sites with shared structure.

New offline RL method handles average-reward MDPs with single-policy coverage.

Study proposes a new risk measure for optimal portfolio allocation.

Pessimistic estimator improves multi-objective policy optimization.

Pessimistic Q-learning improves sample efficiency in offline reinforcement learning.

Optimistic initialisation is an effective strategy for efficient exploration in reinforcement learning (RL). In the tabular case, all provably efficient model-free algorithms rely on it. However, model-free deep RL algorithms do not use optimistic initialisation despite taking inspiration from these provably efficient …

New algorithm reduces best-in-class regret in contextual bandits.

Semi-pessimistic RL tackles distributional shift and data scarcity in offline RL.

PESCAL uses mediators to learn from confounded offline data.

Pessimistic Minimax Value Iteration finds efficient NE policies from offline data.

This paper develops a model of reference-dependent assessment of subjective beliefs in which loss-averse people optimally choose the expectation as the reference point to balance the current felicity from the optimistic anticipation and the future disappointment from the realisation. The choice of over-optimism or over…

Paper develops neural network approximation for pessimistic offline RL with theoretical guarantees.

PASTA optimizes assortment selection using pessimism principle.

Study optimal product assortment using historical data, proving item coverage suffices.

Papers learn from data to make decisions without interacting, improving on previous methods.

MOReL learns offline RL policies using pessimistic MDPs.

Portfolio allocation is crucial for investment companies. However, getting the best strategy in a complex and dynamic stock market is challenging. In this paper, we propose a novel Adaptive Deep Deterministic Reinforcement Learning scheme (Adaptive DDPG) for the portfolio allocation task, which incorporates optimistic …

Study tight offline learning bounds for linear MDPs using variance information.

Investor expectations shifted pessimistically during the 2020 stock market crash and recovery.

This paper is concerned with offline reinforcement learning (RL), which learns using pre-collected data without further exploration. Effective offline RL would be able to accommodate distribution shift and limited data coverage. However, prior algorithms or analyses either suffer from suboptimal sample complexities or …

Develops methods for dynamic pricing in incomplete data settings.

Regularization methods are often employed in deep learning neural networks (DNNs) to prevent overfitting. For penalty based DNN regularization methods, convex penalties are typically considered because of their optimization guarantees. Recent theoretical work have shown that nonconvex penalties that satisfy certain reg…

FedLCB-Q learns optimal policies from federated offline data with linear speedup.

One-bit measurements widely exist in the real world, and they can be used to recover sparse signals. This task is known as the problem of learning halfspaces in learning theory and one-bit compressive sensing (1bit-CS) in signal processing. In this paper, we propose novel algorithms based on both convex and nonconvex s…

New algorithm for reinforcement learning in uncertain environments with unknown thresholds.

The paper studies robust risk measures with linear penalties under uncertain distributions.

The use of machine-learning in neuroimaging offers new perspectives in early diagnosis and prognosis of brain diseases. Although such multivariate methods can capture complex relationships in the data, traditional approaches provide irregular (l2 penalty) or scattered (l1 penalty) predictive pattern with a very limited…

The paper analyzes RLHF with human feedback and provides convergence results for MLE and pessimistic MLE.

We propose a straightforward extension of our previously proposed log-periodic power law model of the ``anti-bubble'' regime of the USA market since the summer of 2000, in terms of the renormalization group framework to model critical points. Using a previous work by Gluzman and Sornette (2002) on the classification of…

New sparse penalty improves biclustering for gene expression data.

New approach avoids excess empirical risk in domain generalization.

We study the problem of estimating high-dimensional regression models regularized by a structured sparsity-inducing penalty that encodes prior structural information on either the input or output variables. We consider two widely adopted types of penalties of this kind as motivating examples: (1) the general overlappin…

A popular heuristic for improved performance in Generative adversarial networks (GANs) is to use some form of gradient penalty on the discriminator. This gradient penalty was originally motivated by a Wasserstein distance formulation. However, the use of gradient penalty in other GAN formulations is not well motivated.…

Curvature penalties improve interpretability of KANs without sacrificing accuracy.

New method reduces bias in sparse Bayesian learning.

We consider a one-period Kyle (1985) framework where the insider can be subject to a penalty if she trades. We establish existence and uniqueness of equilibrium for virtually any penalty function when noise is uniform. In equilibrium, the demand of the insider and the price functions are in general non-linear and remai…

New nonconvex penalty smooths at origin for deep learning.

This work improves policy evaluation and selection using logarithmic smoothing for pessimistic off-policy estimation.

We study the problem of learning high dimensional regression models regularized by a structured-sparsity-inducing penalty that encodes prior structural information on either input or output sides. We consider two widely adopted types of such penalties as our motivating examples: 1) overlapping group lasso penalty, base…

Efficient offline reinforcement learning with neural networks using differentiable function approximation.

Wasserstein GAN(WGAN) is a model that minimizes the Wasserstein distance between a data distribution and sample distribution. Recent studies have proposed stabilizing the training process for the WGAN and implementing the Lipschitz constraint. In this study, we prove the local stability of optimizing the simple gradien…

Multi-group learners suffer a penalty in transductive learning.