# Reinforcement Learning and Adaptive Sampling for Optimized DNN Compilation

arXiv: https://arxiv.org/abs/1905.12799 · HTML: https://mlsanity.com/paper/1905.12799/

## Key facts

- Cited by 9 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 30 May 2019.

## Built on by (most cited first, 9 of 9)

- [A Survey of Machine Learning for Computer Architecture and Systems](https://mlsanity.com/paper/2102.07952/) (2021)
- [Mind Mappings: Enabling Efficient Algorithm-Accelerator Mapping Space Search](https://mlsanity.com/paper/2103.01489/) (2021)
- [FALCON: Feedback-driven Adaptive Long/short-term memory reinforced Coding Optimization system](https://mlsanity.com/paper/2410.21349/) (2024)
- [Customized Monte Carlo Tree Search for LLVM/Polly's Composable Loop Optimization Transformations](https://mlsanity.com/paper/2105.04555/) (2021)
- [AIRCHITECT: Learning Custom Architecture Design and Mapping Space](https://mlsanity.com/paper/2108.08295/) (2021)
- [Demystifying Map Space Exploration for NPUs](https://mlsanity.com/paper/2210.03731/) (2022)
- [Continual Learning Approach for Improving the Data and Computation Mapping in Near-Memory Processing System](https://mlsanity.com/paper/2104.13671/) (2021)
- [ReLATE: Accelerating Tensor Decomposition via Safe and Efficient Learning of Sparse Encodings](https://mlsanity.com/paper/2509.00280/) (2025)
- [ParaCodex: A Profiling-Guided Autonomous Coding Agent for Reliable Parallel Code Generation and Translation](https://mlsanity.com/paper/2601.04327/) (2026)

## Abstract (arXiv)

Achieving faster execution with shorter compilation time can enable further diversity and innovation in neural networks. However, the current paradigm of executing neural networks either relies on hand-optimized libraries, traditional compilation heuristics, or very recently, simulated annealing and genetic algorithms. Our work takes a unique approach by formulating compiler optimizations for neural networks as a reinforcement learning problem, whose solution takes fewer steps to converge. This solution, dubbed ReLeASE, comes with a sampling algorithm that leverages clustering to focus the costly samples (real hardware measurements) on representative points, subsuming an entire subspace. Our adaptive sampling not only reduces the number of samples, but also improves the quality of samples for better exploration in shorter time. As such, experimentation with real hardware shows that reinforcement learning with adaptive sampling provides 4.45x speed up in optimization time over AutoTVM, while also improving inference time of the modern deep networks by 5.6%. Further experiments also confirm that our adaptive sampling can even improve AutoTVM's simulated annealing by 4.00x.