English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Over the last decade, the long-running endeavour to automate high-levelprocesses in machine learning (ML) has risen to mainstream prominence,stimulated by advances in optimisation techniques and their impact on selectingML models/algorithms. Central to this drive is the appeal of engineering acomputational system that both discovers and deploys high-performance solutionsto arbitrary ML problems with minimal human interaction. Beyond this, an evenloftier goal is the pursuit of autonomy, which describes the capability of thesystem to independently adjust an ML solution over a lifetime of changingcontexts. However, these ambitions are unlikely to be achieved in a robustmanner without the broader synthesis of various mechanisms and theoreticalframeworks, which, at the present time, remain scattered across numerousresearch threads. Accordingly, this review seeks to motivate a more expansiveperspective on what constitutes an automated/autonomous ML system, alongsideconsideration of how best to consolidate those elements. In doing so, we surveydevelopments in the following research areas: hyperparameter optimisation,multi-component models, neural architecture search, automated featureengineering, meta-learning, multi-level ensembling, dynamic adaptation,multi-objective evaluation, resource constraints, flexible user involvement,and the principles of generalisation. We also develop a conceptual frameworkthroughout the review, augmented by each topic, to illustrate one possible wayof fusing high-level mechanisms into an autonomous ML system. Ultimately, weconclude that the notion of architectural integration deserves more discussion,without which the field of automated ML risks stifling both its technicaladvantages and general uptake.

AutonoML: Towards an Integrated Framework for Autonomous Machine Learning