Meta-AF: Meta-Learning for Adaptive Filters

Abstract

Adaptive filtering algorithms are pervasive throughout modern society and have had a significant impact on a wide variety of domains including audio processing, telecommunications, biomedical sensing, astropyhysics and cosmology, seismology, and many more. Adaptive filters typically operate via specialized online, iterative optimization methods such as least-mean squares or recursive least squares and aim to process signals in unknown or nonstationary environments. Such algorithms, however, can be slow and laborious to develop, require domain expertise to create, and necessitate mathematical insight for improvement. In this work, we seek to go beyond the limits of human-derived adaptive filter algorithms and present a comprehensive framework for learning online, adaptive signal processing algorithms or update rules directly from data. To do so, we frame the development of adaptive filters as a metalearning problem in the context of deep learning and use a form of self-supervision to learn online iterative update rules for adaptive filters. To demonstrate our approach, we focus on audio applications and systematically develop meta-learned adaptive filters for five canonical audio problems including system identification, acoustic echo cancellation, blind equalization, multichannel dereverberation, and beamforming. For each application, we compare against common baselines and/or current state-ofthe-art methods and show we can learn high-performing adaptive filters that operate in real-time and, in most cases, significantly out perform all past specially developed methods for each task using a single general-purpose configuration of our method.

The Dynamic Problem of an Elastically Supported Beam Analysis Using Neural Architecture Search With Automatic Differentiation

“…In particular, and given a neural architecture search space 𝐹, where the input data 𝐷 is divided into Dtrain and Dval and the cost function Cost(•) (e.g., accuracy, mean squared error, etc. ), the goal is to find an optimal neural network f* ∈ F that can achieve the lowest cost on the dataset D. Finding the optimal neural network f* is equivalent to [12,24,26]:…”

A Hypersensitive Intelligent Filter for Detecting Explicit Content in Learning Environments

“…To tackle this challenge, conventional methods of content filtering and moderation have traditionally relied on rule-based or keyword-based filters [4,5]. Nevertheless, these methods often struggle to accurately detect explicit content due to its ever-evolving nature and nuanced characteristics.…”