ArchGym: An Open-Source Gymnasium for Machine Learning Assisted Architecture Design

by   Srivatsan Krishnan, et al.

Machine learning is a prevalent approach to tame the complexity of design space exploration for domain-specific architectures. Using ML for design space exploration poses challenges. First, it's not straightforward to identify the suitable algorithm from an increasing pool of ML methods. Second, assessing the trade-offs between performance and sample efficiency across these methods is inconclusive. Finally, lack of a holistic framework for fair, reproducible, and objective comparison across these methods hinders progress of adopting ML-aided architecture design space exploration and impedes creating repeatable artifacts. To mitigate these challenges, we introduce ArchGym, an open-source gym and easy-to-extend framework that connects diverse search algorithms to architecture simulators. To demonstrate utility, we evaluate ArchGym across multiple vanilla and domain-specific search algorithms in designing custom memory controller, deep neural network accelerators, and custom SoC for AR/VR workloads, encompassing over 21K experiments. Results suggest that with unlimited samples, ML algorithms are equally favorable to meet user-defined target specification if hyperparameters are tuned; no solution is necessarily better than another (e.g., reinforcement learning vs. Bayesian methods). We coin the term hyperparameter lottery to describe the chance for a search algorithm to find an optimal design provided meticulously selected hyperparameters. The ease of data collection and aggregation in ArchGym facilitates research in ML-aided architecture design space exploration. As a case study, we show this advantage by developing a proxy cost model with an RMSE of 0.61 data for ArchGym is available at


page 1

page 2

page 3

page 4


An Open-Source ML-Based Full-Stack Optimization Framework for Machine Learning Accelerators

Parameterizable machine learning (ML) accelerators are the product of re...

A Survey on Domain-Specific Memory Architectures

The never-ending demand for high performance and energy efficiency is pu...

AIRCHITECT: Learning Custom Architecture Design and Mapping Space

Design space exploration is an important but costly step involved in the...

HDTorch: Accelerating Hyperdimensional Computing with GP-GPUs for Design Space Exploration

HyperDimensional Computing (HDC) as a machine learning paradigm is highl...

Designing for the Long Tail of Machine Learning

Recent technical advances has made machine learning (ML) a promising com...

Apollo: Transferable Architecture Exploration

The looming end of Moore's Law and ascending use of deep learning drives...

Holistic Outpost Design for Lunar Lava Tubes

As the space industry continues its rapid development, humanity is poise...

Please sign up or login with your details

Forgot password? Click here to reset