arXiv research

Improves posterior approximation speed for Dirichlet process mixture models.

Introduces a new theoretical framework for exponential smoothing.

In this paper we focus on the problem of Online Principal Component Analysis in the regret minimization framework. For this problem, all existing regret minimization algorithms for the fully-adversarial setting are based on a positive semidefinite convex relaxation, and hence require quadratic memory and SVD computatio…

Optimizes experimental designs for intractable models using mutual information bounds.

A new acquisition function RMES improves Bayesian optimization performance.

SMAVE optimizes SDR by projecting onto a low-dimensional subspace on a Riemannian manifold.

Studying a softmax-attention model, we show that the learned query converges to the latent signal subspace spanned by the informative direction.

A scalable method for BED with implicit models using approximate gradients.

Gradient ascent method successfully removes specific data points from neural networks without retraining.

Improves BED scalability for implicit models.

Proposes a method to stabilize Black Box Variational Inference using the James-Stein estimator.

We consider parallel global optimization of derivative-free expensive-to-evaluate functions, and propose an efficient method based on stochastic approximation for implementing a conceptual Bayesian optimization algorithm proposed by Ginsbourger et al. (2007). At the heart of this algorithm is maximizing the information…

Study shows how adaptive market agents can lead to persistent overpricing in financial markets.

Policy gradient aims to maximize expected return using gradient ascent.

Unified approach for Bayesian optimal experiment design using stochastic gradients.

Paper introduces SGA for barycenter optimization in optimal transport.

Lazy neural networks are vulnerable to adversarial attacks.

Novel approach simplifies VI problems with faster performance.

We present a new algorithm based on an gradient ascent for a general Active Exploration bandit problem in the fixed confidence setting. This problem encompasses several well studied problems such that the Best Arm Identification or Thresholding Bandits. It consists of a new sampling rule based on an online lazy mirror …

Optimizes binary regression models with gradient ascent-descent methods.

Stochastic optimization of continuous objectives is at the heart of modern machine learning. However, many important problems are of discrete nature and often involve submodular objectives. We seek to unleash the power of stochastic continuous optimization, namely stochastic gradient descent and its variants, to such d…

This paper introduces a new method to evaluate multiple policies simultaneously.

SAVO actor improves reinforcement learning by avoiding local optima in complex Q-functions.

The paper connects two clustering methods by showing gradient ascent flow can move up the cluster tree.

New algorithm solves federated minimax optimization problems.

Algorithm optimizes system design and control for better rewards.

Flooding prevents deep networks from achieving zero training loss, improving test performance.

New method uses deep learning to solve linear inverse problems.

Generalizes PCA to maximize any convex function of components.

RSGDA improves convergence rates for nonconvex-strongly concave optimization.

We comment on the fact that gradient ascent for logistic regression has a connection with the perceptron learning algorithm. Logistic learning is the "soft" variant of perceptron learning.

Method infers parameters in complex diffusion processes.

Score-based martingale posteriors improve uncertainty quantification in deep neural networks.

Paper analyzes adversarial dynamics in neural networks.

Semi-implicit variational inference (SIVI) is introduced to expand the commonly used analytic variational distribution family, by mixing the variational parameter with a flexible distribution. This mixing distribution can assume any density function, explicit or not, as long as independent random samples can be generat…

We present a model for random simple graphs with a degree distribution that obeys a power law (i.e., is heavy-tailed). To attain this behavior, the edge probabilities in the graph are constructed from Bertoin-Fujita-Roynette-Yor (BFRY) random variables, which have been recently utilized in Bayesian statistics for the c…

In this paper we target the class of modal clustering methods where clusters are defined in terms of the local modes of the probability density function which generates the data. The most well-known modal clustering method is the k-means clustering. Mean Shift clustering is a generalization of the k-means clustering wh…

NeuralRBMLE tackles explore-exploit trade-offs in contextual bandits with neural networks.

Multi-subject fMRI data analysis is an interesting and challenging problem in human brain decoding studies. The inherent anatomical and functional variability across subjects make it necessary to do both anatomical and functional alignment before classification analysis. Besides, when it comes to big data, time complex…

Optimal multistage method solves noisy minimax problems.

Gradient methods converge exponentially in concave network games.

Given data with noisy labels, over-parameterized deep networks can gradually memorize the data, and fit everything in the end. Although equipped with corrections for noisy labels, many learning methods in this area still suffer overfitting due to undesired memorization. In this paper, to relieve this issue, we propose …

Develops an MS-inspired algorithm for regression mode finding and space partitioning.

A new method for distributed optimization reduces communication rounds without minibatches.

New method optimizes portfolio weights as functions, outperforming traditional approaches.

Combines neural networks and decision trees for supervised learning.

We investigate a family of poisoning attacks against Support Vector Machines (SVM). Such attacks inject specially crafted training data that increases the SVM's test error. Central to the motivation for these attacks is the fact that most learning algorithms assume that their training data comes from a natural or well-…

Implicit generative models are difficult to train as no explicit density functions are defined. Generative adversarial nets (GANs) present a minimax framework to train such models, which however can suffer from mode collapse due to the nature of the JS-divergence. This paper presents a learning by teaching (LBT) approa…