We gratefully acknowledge support from
the Simons Foundation and member institutions.
Full-text links:

Download:

Current browse context:

stat.ML

Change to browse by:

References & Citations

Bookmark

(what is this?)
CiteULike logo BibSonomy logo Mendeley logo del.icio.us logo Digg logo Reddit logo ScienceWISE logo

Computer Science > Machine Learning

Title: Evolutionary-Neural Hybrid Agents for Architecture Search

Abstract: Neural Architecture Search has shown potential to automate the design of neural networks. Deep Reinforcement Learning based agents can learn complex architectural patterns, as well as explore a vast and compositional search space. On the other hand, evolutionary algorithms offer higher sample efficiency, which is critical for such a resource intensive application. In order to capture the best of both worlds, we propose a class of Evolutionary-Neural hybrid agents (Evo-NAS). We show that the Evo-NAS agent outperforms both neural and evolutionary agents when applied to architecture search for a suite of text and image classification benchmarks. On a high-complexity architecture search space for image classification, the Evo-NAS agent surpasses the accuracy achieved by commonly used agents with only 1/3 of the search cost.
Subjects: Machine Learning (cs.LG); Neural and Evolutionary Computing (cs.NE); Machine Learning (stat.ML)
Cite as: arXiv:1811.09828 [cs.LG]
  (or arXiv:1811.09828v4 [cs.LG] for this version)

Submission history

From: Krzysztof Maziarz [view email]
[v1] Sat, 24 Nov 2018 13:00:47 GMT (787kb,D)
[v2] Thu, 24 Jan 2019 16:05:51 GMT (3029kb,D)
[v3] Tue, 4 Jun 2019 15:27:38 GMT (4945kb,D)
[v4] Sat, 15 Feb 2020 13:25:42 GMT (1535kb,D)

Link back to: arXiv, form interface, contact.