# AIRCHITECT: Learning Custom Architecture Design and Mapping Space

arXiv: https://arxiv.org/abs/2108.08295 · HTML: https://mlsanity.com/paper/2108.08295/

## Key facts

- Cited by 4 papers in the map (data through 14 Oct 2026).
- Builds on 14 of its 33 references that are in the map.
- First posted on arXiv 16 Aug 2021.

## Builds on (most cited first, 14 of 14)

- [Deep Residual Learning for Image Recognition](https://mlsanity.com/paper/1512.03385/) (2015)
- [Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks](https://mlsanity.com/paper/1506.01497/) (2015)
- [XGBoost: A Scalable Tree Boosting System](https://mlsanity.com/paper/1603.02754/) (2016)
- [Going Deeper with Convolutions](https://mlsanity.com/paper/1409.4842/) (2014)
- [MobileNets: Efficient Convolutional Neural Networks for Mobile Vision Applications](https://mlsanity.com/paper/1704.04861/) (2017)
- [In-Datacenter Performance Analysis of a Tensor Processing Unit](https://mlsanity.com/paper/1704.04760/) (2017)
- [Deep Learning Recommendation Model for Personalization and Recommendation Systems](https://mlsanity.com/paper/1906.00091/) (2019)
- [Learning to Optimize Tensor Programs](https://mlsanity.com/paper/1805.08166/) (2018)
- [GCN-RL Circuit Designer: Transferable Transistor Sizing with Graph Neural Networks and Reinforcement Learning](https://mlsanity.com/paper/2005.00406/) (2020)
- [ConfuciuX: Autonomous Hardware Resource Assignment for DNN Accelerators using Reinforcement Learning](https://mlsanity.com/paper/2009.02010/) (2020)
- [Centaur: A Chiplet-based, Hybrid Sparse-Dense Accelerator for Personalized Recommendations](https://mlsanity.com/paper/2005.05968/) (2020)
- [Apollo: Transferable Architecture Exploration](https://mlsanity.com/paper/2102.01723/) (2021)
- [Reinforcement Learning and Adaptive Sampling for Optimized DNN Compilation](https://mlsanity.com/paper/1905.12799/) (2019)
- [NVCell: Standard Cell Layout in Advanced Technology Nodes with Reinforcement Learning](https://mlsanity.com/paper/2107.07044/) (2021)

## Built on by (most cited first, 4 of 4)

- [ArchGym: An Open-Source Gymnasium for Machine Learning Assisted Architecture Design](https://mlsanity.com/paper/2306.08888/) (2023)
- [GANDSE: Generative Adversarial Network based Design Space Exploration for Neural Network Accelerator Design](https://mlsanity.com/paper/2208.00800/) (2022)
- [AIRCHITECT v2: Learning the Hardware Accelerator Design Space through Unified Representations](https://mlsanity.com/paper/2501.09954/) (2025)
- [PerfReasoning: How Well Do LLMs Reason on Hardware Performance?](https://mlsanity.com/paper/2609.04476/) (2026)

## Abstract (arXiv)

Design space exploration is an important but costly step involved in the design/deployment of custom architectures to squeeze out maximum possible performance and energy efficiency. Conventionally, optimizations require iterative sampling of the design space using simulation or heuristic tools. In this paper we investigate the possibility of learning the optimization task using machine learning and hence using the learnt model to predict optimal parameters for the design and mapping space of custom architectures, bypassing any exploration step. We use three case studies involving the optimal array design, SRAM buffer sizing, mapping, and schedule determination for systolic-array-based custom architecture design and mapping space. Within the purview of these case studies, we show that it is possible to capture the design space and train a model to "generalize" prediction the optimal design and mapping parameters when queried with workload and design constraints. We perform systematic design-aware and statistical analysis of the optimization space for our case studies and highlight the patterns in the design space. We formulate the architecture design and mapping as a machine learning problem that allows us to leverage existing ML models for training and inference. We design and train a custom network architecture called AIRCHITECT, which is capable of learning the architecture design space with as high as 94.3% test accuracy and predicting optimal configurations which achieve on average (GeoMean) of 99.9% the best possible performance on a test dataset with 105 GEMM workloads.