Spectral entry-wise matrix estimation for low-rank reinforcement learning
S Stojanovic, Y Jedra… - Advances in Neural …, 2023 - proceedings.neurips.cc
We study matrix estimation problems arising in reinforcement learning with low-rank
structure. In low-rank bandits, the matrix to be recovered specifies the expected arm …
structure. In low-rank bandits, the matrix to be recovered specifies the expected arm …
Shapley meets uniform: An axiomatic framework for attribution in online advertising
One of the central challenges in online advertising is attribution, namely, assessing the
contribution of individual advertiser actions including emails, display ads and search ads to …
contribution of individual advertiser actions including emails, display ads and search ads to …
CP factor model for dynamic tensors
Observations in various applications are frequently represented as a time series of
multidimensional arrays, called tensor time series, preserving the inherent multidimensional …
multidimensional arrays, called tensor time series, preserving the inherent multidimensional …
Online statistical inference for matrix contextual bandit
Contextual bandit has been widely used for sequential decision-making based on the
current contextual information and historical feedback data. In modern applications, such …
current contextual information and historical feedback data. In modern applications, such …
Optimal high-order tensor svd via tensor-train orthogonal iteration
This paper studies a general framework for high-order tensor SVD. We propose a new
computationally efficient algorithm, tensor-train orthogonal iteration (TTOI), that aims to …
computationally efficient algorithm, tensor-train orthogonal iteration (TTOI), that aims to …
Nearly optimal latent state decoding in block mdps
We consider the problem of model estimation in episodic Block MDPs. In these MDPs, the
decision maker has access to rich observations or contexts generated from a small number …
decision maker has access to rich observations or contexts generated from a small number …
[HTML][HTML] Singular value distribution of dense random matrices with block Markovian dependence
J Sanders, A Van Werde - Stochastic Processes and their Applications, 2023 - Elsevier
A block Markov chain is a Markov chain whose state space can be partitioned into a finite
number of clusters such that the transition probabilities only depend on the clusters. Block …
number of clusters such that the transition probabilities only depend on the clusters. Block …
Sparsity-Constraint Optimization via Splicing Iteration
Z Wang, J Zhu, J Zhu, B Tang, H Lin… - arXiv preprint arXiv …, 2024 - arxiv.org
Sparsity-constraint optimization has wide applicability in signal processing, statistics, and
machine learning. Existing fast algorithms must burdensomely tune parameters, such as the …
machine learning. Existing fast algorithms must burdensomely tune parameters, such as the …
From Self-Attention to Markov Models: Unveiling the Dynamics of Generative Transformers
Modern language models rely on the transformer architecture and attention mechanism to
perform language understanding and text generation. In this work, we study learning a 1 …
perform language understanding and text generation. In this work, we study learning a 1 …
Speed up the cold-start learning in two-sided bandits with many arms
Multi-armed bandit (MAB) algorithms are efficient approaches to reduce the opportunity cost
of online experimentation and are used by companies to find the best product from …
of online experimentation and are used by companies to find the best product from …