arXiv research

New suboptimal algorithm for best subset selection in high-dimensional data.

Practical model building processes are often time-consuming because many different models must be trained and validated. In this paper, we introduce a novel algorithm that can be used for computing the lower and the upper bounds of model validation errors without actually training the model itself. A key idea behind ou…

New algorithm reduces suboptimality in imitation learning to nearly optimal levels.

Paper explores limits of imitation learning in MDPs, setting new suboptimality bounds.

The maximum a posteriori (MAP) configuration of binary variable models with submodular graph-structured energy functions can be found efficiently and exactly by graph cuts. Max-product belief propagation (MP) has been shown to be suboptimal on this class of energy functions by a canonical counterexample where MP conver…

A new measure helps compute suboptimality in entropy-regularized methods.

Two entropy measures quantify suboptimal portfolio performance.

In this paper, we propose and study opportunistic bandits - a new variant of bandits where the regret of pulling a suboptimal arm varies under different environmental conditions, such as network load or produce price. When the load/price is low, so is the cost/regret of pulling a suboptimal arm (e.g., trying a suboptim…

New algorithms tackle robust RL with linear models, revealing unique challenges.

In many platforms, user arrivals exhibit a self-reinforcing behavior: future user arrivals are likely to have preferences similar to users who were satisfied in the past. In other words, arrivals exhibit positive externalities. We study multiarmed bandit (MAB) problems with positive externalities. We show that the self…

New algorithm estimates edge density of random graphs robustly, achieving optimal breakdown point.

Novel IRL method identifies suboptimal medical decisions in ICU data.

Curriculum learning has been successfully used in reinforcement learning to accelerate the learning process, through knowledge transfer between tasks of increasing complexity. Critical tasks, in which suboptimal exploratory actions must be minimized, can benefit from curriculum learning, and its ability to shape explor…

A new algorithm reduces suboptimal arm selection in correlated bandits.

New algorithm reduces MDP regret by accounting for state suboptimality gaps and variance.

Paper studies offline RL with linear approx, focusing on inherent Bellman error.

We develop a robust RL algorithm for off-dynamics environments with improved suboptimality bounds and computational efficiency.

Two algorithms improve Federated RL in diverse environments.

Imitation learning (IL) consists of a set of tools that leverage expert demonstrations to quickly learn policies. However, if the expert is suboptimal, IL can yield policies with inferior performance compared to reinforcement learning (RL). In this paper, we aim to provide an algorithm that combines the best aspects of…

There has been substantial research on sub-linear time approximate algorithms for Maximum Inner Product Search (MIPS). To achieve fast query time, state-of-the-art techniques require significant preprocessing, which can be a burden when the number of subsequent queries is not sufficiently large to amortize the cost. Fu…

Layer-wise preconditioning methods improve neural network optimization and feature learning.

Negative momentum accelerates convergence in minimax games but at a suboptimal rate.

We propose novel first-order stochastic approximation algorithms for canonical correlation analysis (CCA). Algorithms presented are instances of inexact matrix stochastic gradient (MSG) and inexact matrix exponentiated gradient (MEG), and achieve $ε$-suboptimality in the population objective in $\operatorname{poly}(\fr…

Paper explores PG for MCR, finding suboptimal policies but providing bounds.

MAPS and MAPS-SE learn from multiple suboptimal experts to improve policies efficiently.

Co-learning BO improves global optimization with limited samples.

Research aims to improve confidence intervals for RKHS elements in online learning.

Develops a Best-of-Both-Worlds algorithm for linear contextual bandits with Tsallis entropy.

The Lasso is suboptimal in sparse linear regression due to design matrix constraints.

Study quantile multi-armed bandits for identifying the best arm with a specified quantile level.

New algorithm reduces regret by focusing on near-optimal actions.

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

Optimal algorithm for high-dimensional stochastic linear bandits with sparse parameters.

Rotation invariant algorithms fail on sparse problems even with noise.

A new algorithm finds minimizers in dueling optimization with a monotone adversary.

New bounds on adaptivity cost in stochastic optimization.

New algorithm tackles confounding in offline RL for partially observable MDPs.

Paper proves suboptimal convergence rate of last iterate for SGDM.

This paper extends the work in [Suzuki, 1996] and presents an efficient depth-first branch-and-bound algorithm for learning Bayesian network structures, based on the minimum description length (MDL) principle, for a given (consistent) variable ordering. The algorithm exhaustively searches through all network structures…

Enhances reinforcement learning from sparse data.

Improved multilevel scheme for value-at-risk computation.

Several probabilistic models from high-dimensional statistics and machine learning reveal an intriguing --and yet poorly understood-- dichotomy. Either simple local algorithms succeed in estimating the object of interest, or even sophisticated semi-definite programming (SDP) relaxations fail. In order to explore this p…

Spectral algorithms improve under covariate shift with novel weighted techniques.

Fine-grained gap-dependent regret bounds for reinforcement learning.

The exploration mechanism used by a Deep Reinforcement Learning (RL) agent plays a key role in determining its sample efficiency. Thus, improving over random exploration is crucial to solve long-horizon tasks with sparse rewards. We propose to leverage an ensemble of partial solutions as teachers that guide the agent's…

Improved LoRA+ adapts large models more efficiently.

Least Squares Estimators are suboptimal for 5D convex functions.

New study shows exponential lower bound for RL even with constant suboptimality gap.