Výsledky vyhledávání - "Verma, Prakhar"

Report

Plan$\times$RAG: Planning-guided Retrieval Augmented Generation

Autor: Verma, Prakhar, Midigeshi, Sukruta Prakash, Sinha, Gaurav, Solin, Arno, Natarajan, Nagarajan, Sharma, Amit

We introduce Planning-guided Retrieval Augmented Generation (Plan$\times$RAG), a novel framework that augments the \emph{retrieve-then-reason} paradigm of existing RAG frameworks to \emph{plan-then-retrieve}. Plan$\times$RAG formulates a reasoning pl

Externí odkaz: http://arxiv.org/abs/2410.20753

Zobrazit plný text záznamu

Report

Memory-Based Dual Gaussian Processes for Sequential Learning

Autor: Chang, Paul E., Verma, Prakhar, John, S. T., Solin, Arno, Khan, Mohammad Emtiyaz

Sequential learning with Gaussian processes (GPs) is challenging when access to past data is limited, for example, in continual and active learning. In such cases, errors can accumulate over time due to inaccuracies in the posterior, hyperparameters,

Externí odkaz: http://arxiv.org/abs/2306.03566

Zobrazit plný text záznamu

Report

Variational Gaussian Process Diffusion Processes

Autor: Verma, Prakhar, Adam, Vincent, Solin, Arno

Diffusion processes are a class of stochastic differential equations (SDEs) providing a rich family of expressive models that arise naturally in dynamic modelling tasks. Probabilistic inference and learning under generative models with latent process

Externí odkaz: http://arxiv.org/abs/2306.02066

Zobrazit plný text záznamu

Report

PriorCVAE: scalable MCMC parameter inference with Bayesian deep generative modelling

Autor: Semenova, Elizaveta, Verma, Prakhar, Cairney-Leeming, Max, Solin, Arno, Bhatt, Samir, Flaxman, Seth

Recent advances have shown that GP priors, or their finite realisations, can be encoded using deep generative models such as variational autoencoders (VAEs). These learned generators can serve as drop-in replacements for the original priors during MC

Externí odkaz: http://arxiv.org/abs/2304.04307

Zobrazit plný text záznamu

Report

Fantasizing with Dual GPs in Bayesian Optimization and Active Learning

Autor: Chang, Paul E., Verma, Prakhar, John, ST, Picheny, Victor, Moss, Henry, Solin, Arno

Gaussian processes (GPs) are the main surrogate functions used for sequential modelling such as Bayesian Optimization and Active Learning. Their drawbacks are poor scaling with data and the need to run an optimization loop when using a non-Gaussian l

Externí odkaz: http://arxiv.org/abs/2211.01053

Zobrazit plný text záznamu

Report

Scalable Inference in SDEs by Direct Matching of the Fokker-Planck-Kolmogorov Equation

Autor: Solin, Arno, Tamir, Ella, Verma, Prakhar

Simulation-based techniques such as variants of stochastic Runge-Kutta are the de facto approach for inference with stochastic differential equations (SDEs) in machine learning. These methods are general-purpose and used with parametric and non-param

Externí odkaz: http://arxiv.org/abs/2110.15739

Zobrazit plný text záznamu