Learning to Learn to Demodulate with Uncertainty Quantification via Bayesian Meta-Learning

Kfir M. Cohen, Sangwoo Park, Osvaldo Simeone, Shlomo Shamai

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

Meta-learning, or learning to learn, offers a principled framework for few-shot learning. It leverages data from multiple related learning tasks to infer an inductive bias that enables fast adaptation on a new task. The application of meta-learning was recently proposed for learning how to demodulate from few pilots. The idea is to use pilots received and stored for offline use from multiple devices in order to meta-learn an adaptation procedure with the aim of speeding up online training on new devices. Standard frequentist learning, which can yield relatively accurate “hard” classification decisions, is known to be poorly calibrated, particularly in the small-data regime. Poor calibration implies that the soft scores output by the demodulator are inaccurate estimates of the true probability of correct demodulation. In this work, we introduce the use of Bayesian meta-learning via variational inference for the purpose of obtaining well-calibrated few-pilot demodulators. In a Bayesian framework, each neural network weight is represented by a distribution, capturing epistemic uncertainty. Bayesian meta-learning optimizes over the prior distribution of the weights. The resulting Bayesian ensembles offer better calibrated soft decisions, at the computational cost of running multiple instances of the neural network for demodulation. Numerical results for single-input single-output Rayleigh fading channels with transmitter’s non-linearities are provided that compare symbol error rate and expected calibration error for both frequentist and Bayesian meta-learning, illustrating how the latter is both more accurate and better-calibrated.

Original languageEnglish (US)
Title of host publicationWSA 2021 - 25th International ITG Workshop on Smart Antennas
PublisherVDE Verlag GmbH
Pages202-207
Number of pages6
ISBN (Electronic)9783800756889
StatePublished - 2021
Externally publishedYes
Event25th International ITG Workshop on Smart Antennas, WSA 2021 - French Riviera, France
Duration: Nov 10 2021Nov 12 2021

Publication series

NameWSA 2021 - 25th International ITG Workshop on Smart Antennas

Conference

Conference25th International ITG Workshop on Smart Antennas, WSA 2021
Country/TerritoryFrance
CityFrench Riviera
Period11/10/2111/12/21

All Science Journal Classification (ASJC) codes

  • Computer Networks and Communications
  • Instrumentation

Fingerprint

Dive into the research topics of 'Learning to Learn to Demodulate with Uncertainty Quantification via Bayesian Meta-Learning'. Together they form a unique fingerprint.

Cite this