Multi-kernel Passive Stochastic Gradient Algorithms and Transfer Learning

2020-08-23 · Vikram Krishnamurthy, George Yin

This paper develops a novel passive stochastic gradient algorithm. In passive stochastic approximation, the stochastic gradient algorithm does not have control over the location where noisy gradients of the cost function are evaluated. Classical passive stochastic gradient algorithms use a kernel that approximates a Dirac delta to weigh the gradients based on how far they are evaluated from the desired point. In this paper we construct a multi-kernel passive stochastic gradient algorithm. The algorithm performs substantially better in high dimensional problems and incorporates variance reduction. We analyze the weak convergence of the multi-kernel algorithm and its rate of convergence. In numerical examples, we study the multi-kernel version of the passive least mean squares (LMS) algorithm for transfer learning to compare the performance with the classical passive version.

📄 PDF Abstract BibTeX arXiv:2008.10020

Code (0)

등록된 구현이 없습니다.

Tasks

Transfer Learning

Langevin Dynamics for Adaptive Inverse Reinforcement Learning of Stochastic Gradient Algorithms

2020-06-20 · Vikram Krishnamurthy, George Yin

Inverse reinforcement learning (IRL) aims to estimate the reward function of optimizing agents by observing their response (estimates or actions). This paper considers IRL when noisy estimates of the gradient of a reward…

reinforcement-learning Reinforcement Learning (RL)

Finite-Sample Bounds for Adaptive Inverse Reinforcement Learning using Passive Langevin Dynamics

2023-04-18 · Luke Snow, Vikram Krishnamurthy

This paper provides a finite-sample analysis of a passive stochastic gradient Langevin dynamics (PSGLD) algorithm. This algorithm is designed to achieve adaptive inverse reinforcement learning (IRL). Adaptive IRL aims to…

Density Estimation reinforcement-learning Reinforcement Learning

Smoothed Functional Algorithms for Stochastic Optimization using q-Gaussian Distributions

2012-06-21 · Debarghya Ghoshdastidar, Ambedkar Dukkipati, Shalabh Bhatnagar

Smoothed functional (SF) schemes for gradient estimation are known to be efficient in stochastic optimization algorithms, specially when the objective is to improve the performance of a stochastic system. However, the pe…

Stochastic Optimization

Optimal Convergence for Distributed Learning with Stochastic Gradient Methods and Spectral Algorithms

2018-01-22 · Junhong Lin, Volkan Cevher

We study generalization properties of distributed algorithms in the setting of nonparametric regression over a reproducing kernel Hilbert space (RKHS). We first investigate distributed stochastic gradient methods (SGM), …

regression

Efficient Online Relative Comparison Kernel Learning

2015-01-06 · Eric Heim, Matthew Berger, Lee M. Seversky, Milos Hauskrecht

Learning a kernel matrix from relative comparison human feedback is an important problem with applications in collaborative filtering, object retrieval, and search. For learning a kernel over a large number of objects, e…

Collaborative Filtering Retrieval