Searcharxiv⌕ Search

arXiv subjects

Arezou Rezazadeh

Publications and source records attributed to Arezou Rezazadeh.

5 recordsLinked to original sources
arXiv2022

A General framework for PAC-Bayes Bounds for Meta-Learning

Meta learning automatically infers an inductive bias, that includes the hyperparameter of the base-learning algorithm, by observing data from a finite number of related tasks. This paper studies PAC-Bayes bounds on meta generalization gap. The meta-generalization gap comprises two sources of generalization gaps: the environment-level and task-level gaps resulting from observation of a finite number of tasks and data samples per task, respectively. In this paper, by upper bounding arbitrary convex functions, which link the expected and empirical losses at the environment and also per-task levels, we obtain new PAC-Bayes bounds. Using these bounds, we develop new PAC-Bayes meta-learning algorithms. Numerical examples demonstrate the merits of the proposed novel bounds and algorithm in comparison to prior PAC-Bayes bounds for meta-learning.

cs.LG↗
arXiv2021

Conditional Mutual Information-Based Generalization Bound for Meta Learning

Meta-learning optimizes an inductive bias---typically in the form of the hyperparameters of a base-learning algorithm---by observing data from a finite number of related tasks. This paper presents an information-theoretic bound on the generalization performance of any given meta-learner, which builds on the conditional mutual information (CMI) framework of Steinke and Zakynthinou (2020). In the proposed extension to meta-learning, the CMI bound involves a training \textit{meta-supersample} obtained by first sampling $2N$ independent tasks from the task environment, and then drawing $2M$ independent training samples for each sampled task. The meta-training data fed to the meta-learner is modelled as being obtained by randomly selecting $N$ tasks from the available $2N$ tasks and $M$ training samples per task from the available $2M$ training samples per task. The resulting bound is explicit in two CMI terms, which measure the information that the meta-learner output and the base-learner output provide about which training data are selected, given the entire meta-supersample. Finally, we present a numerical example that illustrates the merits of the proposed bound in comparison to prior information-theoretic bounds for meta-learning.

cs.LG↗
arXiv2019

Joint Source-Channel Coding for the Multiple-Access Channel with Correlated Sources

This paper studies the random-coding exponent of joint source-channel coding for the multiple-access channel with correlated sources. For each user, by defining a threshold, the messages of each source are partitioned into two classes. The achievable exponent for correlated sources with two message-dependent input distributions for each user is determined and shown to be larger than that achieved using only one input distribution for each user. A system of equations is presented to determine the optimal thresholds maximizing the achievable exponent. The obtained exponent is compared with the one derived for the MAC with independent sources.

cs.IT↗
arXiv2018

Multiple-Access Channel with Independent Sources: Error Exponent Analysis

In this paper, an achievable error exponent for the multiple-access channel with two independent sources is derived. For each user, the source messages are partitioned into two classes and codebooks are generated by drawing codewords from an input distribution depending on the class index of the source message. The partitioning thresholds that maximize the achievable exponent are given by the solution of a system of equations. We also derive both lower and upper bounds for the achievable exponent in terms of Gallager's source and channel functions. Finally, a numerical example shows that using the proposed ensemble gives a noticeable gain in terms of exponent with respect to independent identically distributed codebooks.

cs.IT↗