arXiv research

Model learns continuously, starting from a blank slate.

New algorithms reduce slate bandit regret for large slates, outperforming existing methods.

Generative model learns to create coherent slates from prompts.

Scalable model for slate recommendation learns reward probabilities.

A deep reinforcement learning approach for slate re-ranking in e-commerce.

A new estimator reduces variance in slate bandit OPE.

The conventional solution to the recommendation problem greedily ranks individual document candidates by prediction scores. However, this method fails to optimize the slate as a whole, and hence, often struggles to capture biases caused by the page layout and document interdepedencies. The slate recommendation problem …

A new recommender system uses slates and Thompson Sampling to improve diversity and click rates.

We learn hierarchical slate representations for collaborative filtering.

This work tackles slate-based recommender systems using RL, optimizing long-term user engagement.

Efficient algorithms for contextual slate bandits with limited adaptivity.

New estimator reduces risk in slate bandits by leveraging Bayes risk criterion.

Deep neural nets classify tool wear in blanking processes.

MaskGAN fills in text blanks more flexibly and effectively.

New algorithms learn MNL weights efficiently for any slate size.

This paper optimizes slate decision systems for large action spaces.

Ranking is a central task in machine learning and information retrieval. In this task, it is especially important to present the user with a slate of items that is appealing as a whole. This in turn requires taking into account interactions between items, since intuitively, placing an item on the slate affects the deci…

A new dataset tracks user interactions and click responses in online marketplaces.

A new algorithm THV-UCB reduces regret in multi-objective bandit problems.

This paper studies the evaluation of policies that recommend an ordered set of items (e.g., a ranking) based on some context---a common scenario in web search, ads, and recommendation. We build on techniques from combinatorial bandits to introduce a new practical estimator that uses logged data to estimate a policy's p…

Improved speech recognition model with better performance.

New method for evaluating sequential recommendations with lower variance.

New tiles allow efficient knot mosaics for small knots.

Unified framework for geometric computation of minimum-area homotopy.

We study the problem to extend an immersed circle f in the 2-dimensional sphere to an immersion of the disc. We analyze existence and uniqueness for this problems in terms of the combinatorial structure of a word assigned to f. Our techniques are based on ideas of Blank who studied the extension problem in case of a pl…

Tensor networks improve sequence modeling with efficient sampling and parallelism.

Hybrid methods that utilize both content and rating information are commonly used in many recommender systems. However, most of them use either handcrafted features or the bag-of-words representation as a surrogate for the content information but they are neither effective nor natural enough. To address this problem, w…

New method classifies immersed surfaces extending from circles in surfaces.

We summarize the potential impact that the European Union's new General Data Protection Regulation will have on the routine use of machine learning algorithms. Slated to take effect as law across the EU in 2018, it will restrict automated individual decision-making (that is, algorithms that make decisions based on user…

Study explains Zipf's law using geometric mechanisms from a finite alphabet.

In this paper we introduce the concept of a space-efficient knot mosaic. That is, we seek to determine how to create knot mosaics using the least number of non-blank tiles necessary to depict the knot. This least number is called the tile number of the knot. We determine strict bounds for the tile number of a knot in t…

In 2008, Kauffman and Lomonaco introduce the concepts of a knot mosaic and the mosaic number of a knot or link, the smallest integer $n$ such that a knot or link can be represented on an $n$-mosaic. In arXiv:1702.06462, the authors explore space-efficient knot mosaics and the tile number of a knot or link, the smallest…

Researchers show mixtures of ranking models are generally identifiable.

Algorithm finds mosaic numbers for knots with 10 or fewer crossings.

Paper solves Minkowski problem for q-torsional rigidity using curvature flow.

New group found not satisfying quasi-isometric triviality property.

Streamable model improves speech recognition performance.

Machine learning algorithms such as linear regression, SVM and neural network have played an increasingly important role in the process of scientific discovery. However, none of them is both interpretable and accurate on nonlinear datasets. Here we present contextual regression, a method that joins these two desirable …

Study examines biases in clinical word embeddings, revealing performance gaps across groups.

This contribution deals with the generalized symmetric FastICA algorithm in the domain of Independent Component Analysis (ICA). The generalized symmetric version of FastICA has been shown to have the potential to achieve the Cramér-Rao Bound (CRB) by allowing the usage of different nonlinearity functions in its paralle…

This paper analyzes stablecoins to address cryptocurrency volatility.

RL agents optimize only specified features; this project infers unmentioned preferences from the state of the environment.

Proposes Genetic Thompson Sampling for multi-armed bandits, improving performance in nonstationary settings.

PoPCoin aims to create a more equitable cryptocurrency.

Measuring the corporate default risk is broadly important in economics and finance. Quantitative methods have been developed to predictively assess future corporate default probabilities. However, as a more difficult yet crucial problem, evaluating the uncertainties associated with the default predictions remains littl…

This paper solves the dual Minkowski problem for q-torsional rigidity.

Enhances currency strategy Sharpe ratio by 30% using context-aware Learning to Rank.

In this paper we perform a blow-up and quantization analysis of the fractional Liouville equation in dimension $1$. More precisely, given a sequence $u_k :\mathbb{R} \to \mathbb{R}$ of solutions to \begin{equation} (-Δ)^\frac{1}{2} u_k =K_ke^{u_k}\quad \text{in }\mathbb{R}, \end{equation} with $K_k$ bounded in $L^\inft…