arXiv research

Optimizes sample and round complexity in adaptive sampling from multiple distributions.

We consider the problem of learning a general graph $G=(V,E)$ using edge-detecting queries, where the number of vertices $|V|=n$ is given to the learner. The information theoretic lower bound gives $m\log n$ for the number of queries, where $m=|E|$ is the number of edges. In case the number of edges $m$ is also given t…

Two new algorithms optimize decentralized convex optimization with reduced communication rounds.

A new algorithm reduces communication rounds for distributed convex optimization.

New algorithms for streaming bandits with limited memory.

Unified framework for combinatorial and rounding algorithms in experimental design.

SS-SARSA tackles recovering bandits by treating rounds as states.

New algorithm reduces online learning regret for bounded recall games.

We study distributed optimization algorithms for minimizing the average of convex functions. The applications include empirical risk minimization problems in statistical machine learning where the datasets are large and have to be stored on different machines. We design a distributed stochastic variance reduced gradien…

This paper considers the multi-armed thresholding bandit problem -- identifying all arms whose expected rewards are above a predefined threshold via as few pulls (or rounds) as possible -- proposed by Locatelli et al. [2016] recently. Although the proposed algorithm in Locatelli et al. [2016] achieves the optimal round…

We present a parallel algorithm that computes the ask and bid prices of an American option when proportional transaction costs apply to the trading of the underlying asset. The algorithm computes the prices on recombining binomial trees, and is designed for modern multi-core processors. Although parallel option pricing…

We propose a new algorithm for adversarial multi-armed bandits with unrestricted delays. The algorithm is based on a novel hybrid regularizer applied in the Follow the Regularized Leader (FTRL) framework. It achieves $\mathcal{O}(\sqrt{kn}+\sqrt{D\log(k)})$ regret guarantee, where $k$ is the number of arms, $n$ is the …

A new reinforcement learning method improves Max-Cut solutions without needing training data.

New algorithm achieves near-optimal performance in dueling bandit problem.

Most traditional online learning algorithms are based on variants of mirror descent or follow-the-leader. In this paper, we present an online algorithm based on a completely different approach, tailored for transductive settings, which combines "random playout" and randomized rounding of loss subgradients. As an applic…

A practical one-shot federated learning algorithm for cross-silo setting.

Advice-efficient prediction with expert advice (in analogy to label-efficient prediction) is a variant of prediction with expert advice game, where on each round of the game we are allowed to ask for advice of a limited number $M$ out of $N$ experts. This setting is especially interesting when asking for advice of ever…

We introduce a model-free relax-and-round algorithm for k-means clustering based on a semidefinite relaxation due to Peng and Wei. The algorithm interprets the SDP output as a denoised version of the original data and then rounds this output to a hard clustering. We provide a generic method for proving performance guar…

Two algorithms achieve optimal regret with limited adaptivity in multinomial logistic bandits.

A new decentralized algorithm DESTINY solves optimization over Stiefel manifold with single communication round.

Improved algorithm reduces communication rounds for distributed online learning.

We present an adaptive online gradient descent algorithm to solve online convex optimization problems with long-term constraints , which are constraints that need to be satisfied when accumulated over a finite number of rounds T , but can be violated in intermediate rounds. For some user-defined trade-off parameter $β$…

One-round FL method improves robustness and reduces communication rounds.

Paper generalizes reward distribution in multi-armed bandits with temporally-partitioned rewards.

We propose a simple yet powerful framework for modeling integer-valued data, such as counts, scores, and rounded data. The data-generating process is defined by Simultaneously Transforming and Rounding (STAR) a continuous-valued process, which produces a flexible family of integer-valued distributions capable of modeli…

CRA improves UL-based CO solvers by dynamically smoothing and enforcing discreteness.

We present and analyze a simple, two-step algorithm to approximate the optimal solution of the sparse PCA problem. Our approach first solves a L1 penalized version of the NP-hard sparse PCA optimization problem and then uses a randomized rounding strategy to sparsify the resulting dense solution. Our main theoretical r…

New predictive bandit model with noise for better decision making.

New algorithm reduces heavy-tailed linear bandits' computational cost.

We present explicit algorithms for simplifying the topology of indefinite fibrations on 4-manifolds, which include broken Lefschetz fibrations and indefinite Morse 2-functions. The algorithms consist of sequences of moves, which modify indefinite fibrations in smooth 1-parameter families. In particular, given an arbitr…

Recently, there has been an increasing interest in designing distributed convex optimization algorithms under the setting where the data matrix is partitioned on features. Algorithms under this setting sometimes have many advantages over those under the setting where data is partitioned on samples, especially when the …

A new TS-SA method alleviates non-stationarity in TS algorithms for bandits.

Improves sampling, rounding, and integration of logconcave functions.

We address the M-best-arm identification problem in multi-armed bandits. A player has a limited budget to explore K arms (M<K), and once pulled, each arm yields a reward drawn (independently) from a fixed, unknown distribution. The goal is to find the top M arms in the sense of expected reward. We develop an algorithm …

New algorithm reduces regret in both adversarial and stochastic contexts.

This paper tackles combinatorial pure exploration for dueling bandits, aiming to find the best candidate-position match.

Algorithm allocates perishable resources online to minimize envy and inefficiency.

ACOWA improves distributed sparse classification with extra communication round.

The paper develops a robust algorithm for contextual bandits with heavy-tailed rewards.

Balances and eliminates base algorithms in bandits and RL to bound total regret.

Recently, reducing communication time between machines becomes the main focus of distributed data mining. Previous methods propose to make workers do more computation locally before aggregating local solutions in the server such that fewer communication rounds between server and workers are required. However, these met…

Gradient descent with biased rounding errors converges faster under certain conditions.

We study the multi-armed bandit problem with subgaussian rewards. The explore-then-commit (ETC) strategy, which consists of an exploration phase followed by an exploitation phase, is one of the most widely used algorithms in a variety of online decision applications. Nevertheless, it has been shown in Garivier et al. (…

In this paper, we study the non-stationary online second price auction problem. We assume that the seller is selling the same type of items in $T$ rounds by the second price auction, and she can set the reserve price in each round. In each round, the bidders draw their private values from a joint distribution unknown t…

New algorithm reduces adaptation lag in online model selection.

The stochastic linear bandit problem proceeds in rounds where at each round the algorithm selects a vector from a decision set after which it receives a noisy linear loss parameterized by an unknown vector. The goal in such a problem is to minimize the (pseudo) regret which is the difference between the total expected …

We study the problem of online learning in a class of Markov decision processes known as linearly solvable MDPs. In the stationary version of this problem, a learner interacts with its environment by directly controlling the state transitions, attempting to balance a fixed state-dependent cost and a certain smooth cost…

Efficient methods reduce projections in non-stationary online learning.