# Data-Driven Offline Optimization For Architecting Hardware Accelerators

arXiv: https://arxiv.org/abs/2110.11346 · HTML: https://mlsanity.com/paper/2110.11346/

## Key facts

- Cited by 29 papers in the map (data through 14 Oct 2026).
- Builds on 38 of its 67 references that are in the map.
- First posted on arXiv 20 Oct 2021.

## Builds on (most cited first, 30 of 38)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [Adam: A Method for Stochastic Optimization](https://mlsanity.com/paper/1412.6980/) (2014)
- [Language Models are Few-Shot Learners](https://mlsanity.com/paper/2005.14165/) (2020)
- [U-Net: Convolutional Networks for Biomedical Image Segmentation](https://mlsanity.com/paper/1505.04597/) (2015)
- [Explaining and Harnessing Adversarial Examples](https://mlsanity.com/paper/1412.6572/) (2014)
- [MobileNetV2: Inverted Residuals and Linear Bottlenecks](https://mlsanity.com/paper/1801.04381/) (2018)
- [Auto-Encoding Variational Bayes](https://mlsanity.com/paper/1312.6114/) (2013)
- [Outrageously Large Neural Networks: The Sparsely-Gated Mixture-of-Experts Layer](https://mlsanity.com/paper/1701.06538/) (2017)
- [Searching for MobileNetV3](https://mlsanity.com/paper/1905.02244/) (2019)
- [Conservative Q-Learning for Offline Reinforcement Learning](https://mlsanity.com/paper/2006.04779/) (2020)
- [In-Datacenter Performance Analysis of a Tensor Processing Unit](https://mlsanity.com/paper/1704.04760/) (2017)
- [Conditional Neural Processes](https://mlsanity.com/paper/1807.01613/) (2018)
- [Neural Processes](https://mlsanity.com/paper/1807.01622/) (2018)
- [Scalable Bayesian Optimization Using Deep Neural Networks](https://mlsanity.com/paper/1502.05700/) (2015)
- [Attentive Neural Processes](https://mlsanity.com/paper/1901.05761/) (2019)
- [Streaming End-to-end Speech Recognition For Mobile Devices](https://mlsanity.com/paper/1811.06621/) (2018)
- [Conditioning by adaptive sampling for robust design](https://mlsanity.com/paper/1901.10060/) (2019)
- [Bayesian Optimization with Unknown Constraints](https://mlsanity.com/paper/1403.5607/) (2014)
- [A Streaming On-Device End-to-End Model Surpassing Server-Side Conventional Model Quality and Latency](https://mlsanity.com/paper/2003.12710/) (2020)
- [Conservative Objective Models for Effective Offline Model-Based Optimization](https://mlsanity.com/paper/2107.06882/) (2021)
- [Preventing Posterior Collapse with delta-VAEs](https://mlsanity.com/paper/1901.03416/) (2019)
- [Model Inversion Networks for Model-Based Optimization](https://mlsanity.com/paper/1912.13464/) (2019)
- [Chip Placement with Deep Reinforcement Learning](https://mlsanity.com/paper/2004.10746/) (2020)
- [A Better and Faster End-to-End Model for Streaming ASR](https://mlsanity.com/paper/2011.10798/) (2020)
- [Autofocused oracles for model-based design](https://mlsanity.com/paper/2006.08052/) (2020)
- [Automatically designing CNN architectures using genetic algorithm for image classification](https://mlsanity.com/paper/1808.03818/) (2018)
- [An Evaluation of Edge TPU Accelerators for Convolutional Neural Networks](https://mlsanity.com/paper/2102.10423/) (2021)
- [Offline Model-Based Optimization via Normalized Maximum Likelihood Estimation](https://mlsanity.com/paper/2102.07970/) (2021)
- [Practical Design Space Exploration](https://mlsanity.com/paper/1810.05236/) (2018)
- [ConfuciuX: Autonomous Hardware Resource Assignment for DNN Accelerators using Reinforcement Learning](https://mlsanity.com/paper/2009.02010/) (2020)

## Built on by (most cited first, 29 of 29)

- [Learning Performance-Improving Code Edits](https://mlsanity.com/paper/2302.07867/) (2023)
- [OmniPred: Language Models as Universal Regressors](https://mlsanity.com/paper/2402.14547/) (2024)
- [Unsupervised Learning for Combinatorial Optimization with Principled Objective Relaxation](https://mlsanity.com/paper/2207.05984/) (2022)
- [Data-Driven Offline Decision-Making via Invariant Representation Learning](https://mlsanity.com/paper/2211.11349/) (2022)
- [Searching for High-Value Molecules Using Reinforcement Learning and Transformers](https://mlsanity.com/paper/2310.02902/) (2023)
- [Offline Model-Based Optimization: Comprehensive Review](https://mlsanity.com/paper/2503.17286/) (2025)
- [Offline Model-Based Optimization by Learning to Rank](https://mlsanity.com/paper/2410.11502/) (2024)
- [Functional Graphical Models: Structure Enables Offline Data-Driven Optimization](https://mlsanity.com/paper/2401.05442/) (2024)
- [GRANITE: A Graph Neural Network Model for Basic Block Throughput Estimation](https://mlsanity.com/paper/2210.03894/) (2022)
- [Adaptive Sampling of k-Space in Magnetic Resonance for Rapid Pathology Prediction](https://mlsanity.com/paper/2406.04318/) (2024)
- [Towards Universal Offline Black-Box Optimization via Learning Language Model Embeddings](https://mlsanity.com/paper/2506.07109/) (2025)
- [ArchGym: An Open-Source Gymnasium for Machine Learning Assisted Architecture Design](https://mlsanity.com/paper/2306.08888/) (2023)
- [Latent Conservative Objective Models for Data-Driven Crystal Structure Prediction](https://mlsanity.com/paper/2310.10056/) (2023)
- [Dynamic Co-Optimization Compiler: Leveraging Multi-Agent Reinforcement Learning for Enhanced DNN Accelerator Performance](https://mlsanity.com/paper/2407.08192/) (2024)
- [Integrated Hardware Architecture and Device Placement Search](https://mlsanity.com/paper/2407.13143/) (2024)
- [DOSA: Differentiable Model-Based One-Loop Search for DNN Accelerators](https://mlsanity.com/paper/2509.10702/) (2025)
- [Cliqueformer: Model-Based Optimization with Structured Transformers](https://mlsanity.com/paper/2410.13106/) (2024)
- [On the Learnability of Offline Model-Based Optimization: A Ranking Perspective](https://mlsanity.com/paper/2603.04000/) (2026)
- [From Function to Distribution Modeling: A PAC-Generative Approach to Offline Optimization](https://mlsanity.com/paper/2401.02019/) (2024)
- [Large Language Model as Meta-Surrogate for Data-Driven Many-Task Optimization: A Proof-of-Principle Study](https://mlsanity.com/paper/2503.08301/) (2025)
- [Learning Design-Score Manifold to Guide Diffusion Models for Offline Optimization](https://mlsanity.com/paper/2506.05680/) (2025)
- [Multi-Agent Reinforcement Learning for Microprocessor Design Space Exploration](https://mlsanity.com/paper/2211.16385/) (2022)
- [DEAP: Design Space Exploration for DNN Accelerator Parallelism](https://mlsanity.com/paper/2312.15388/) (2023)
- [ATraDiff: Accelerating Online Reinforcement Learning with Imaginary Trajectories](https://mlsanity.com/paper/2406.04323/) (2024)
- [Island-Based Evolutionary Computation with Diverse Surrogates and Adaptive Knowledge Transfer for High-Dimensional Data-Driven Optimization](https://mlsanity.com/paper/2503.12856/) (2025)
- [Offline Multi-Task Multi-Objective Data-Driven Evolutionary Algorithm with Language Surrogate Model and Implicit Q-Learning](https://mlsanity.com/paper/2512.15149/) (2025)
- [Offline Materials Optimization with CliqueFlowmer](https://mlsanity.com/paper/2603.06082/) (2026)
- [ChatNeuroSim: An LLM Agent Framework for Automated Compute-in-Memory Accelerator Deployment and Optimization](https://mlsanity.com/paper/2603.08745/) (2026)
- [PerfReasoning: How Well Do LLMs Reason on Hardware Performance?](https://mlsanity.com/paper/2609.04476/) (2026)

## Abstract (arXiv)

Industry has gradually moved towards application-specific hardware accelerators in order to attain higher efficiency. While such a paradigm shift is already starting to show promising results, designers need to spend considerable manual effort and perform a large number of time-consuming simulations to find accelerators that can accelerate multiple target applications while obeying design constraints. Moreover, such a "simulation-driven" approach must be re-run from scratch every time the set of target applications or design constraints change. An alternative paradigm is to use a "data-driven", offline approach that utilizes logged simulation data, to architect hardware accelerators, without needing any form of simulations. Such an approach not only alleviates the need to run time-consuming simulation, but also enables data reuse and applies even when set of target applications changes. In this paper, we develop such a data-driven offline optimization method for designing hardware accelerators, dubbed PRIME, that enjoys all of these properties. Our approach learns a conservative, robust estimate of the desired cost function, utilizes infeasible points, and optimizes the design against this estimate without any additional simulator queries during optimization. PRIME architects accelerators -- tailored towards both single and multiple applications -- improving performance upon state-of-the-art simulation-driven methods by about 1.54x and 1.20x, while considerably reducing the required total simulation time by 93% and 99%, respectively. In addition, PRIME also architects effective accelerators for unseen applications in a zero-shot setting, outperforming simulation-based methods by 1.26x.