Machine Learning-Guided Design Space Exploration for Instruction-Level Approximate Computing

• Gregório K. Neto UFMS
• Marco Antônio de R. Zárate UFMS
• Gislayne G. Damasceno UFMS
• Bruno Nogueira UFMS
• Daniela Catelan UFMS
• Ricardo R. Santos UFMS
• Liana Duenha UFMS

DOI: https://doi.org/10.5753/sscad.2026.34963

Resumo

Approximate Computing (AC) trades reduced precision for energy savings in error-tolerant scenarios, but identifying the right instruction-level placement demands extensive, computationally prohibitive Design Space Exploration (DSE), further burdened by costly simulation-based evaluation. To address this, we propose a heuristic that reduces the design space via an energy-accuracy trade-off, combined with machine learning-based models that replace simulation during DSE. We introduced custom analysis passes to extract program features for these models, and trained predictors achieved R² ≥ 0.955 for error estimation and R² ≥ 0.986 for energy estimation across all benchmarks, with Spearman rank correlations at least ρ = 0.966, confirming that solution ordering is faithfully preserved against full simulation. Replacing simulation with ML-based prediction yielded mean speedups from ≈ 23,987× to ≈ 823,693×, demonstrating that instruction-level approximate DSE can be made tractable without meaningful loss of solution quality.

Referências

Aho, A. V., Lam, M. S., Sethi, R., and Ullman, J. D. (2006). Compilers: Principles, Techniques, and Tools. Pearson Education, Boston, 2nd edition.

Akhlaghi, V., Yazdanbakhsh, A., Samadi, K., Gupta, R. K., and Esmaeilzadeh, H. (2018). Snapea: Predictive early activation for reducing computation in deep convolutional neural networks. In 2018 ACM/IEEE 45th Annual International Symposium on Computer Architecture (ISCA), pages 662–673. IEEE.

Almurib, H. A., Kumar, T. N., and Lombardi, F. (2016). Inexact designs for approximate low power addition by cell replacement. In 2016 Design, Automation & Test in Europe Conference & Exhibition (DATE), pages 660–665. IEEE.

Barros, G., Akemaru, R., Bezerra, E., Oliveira, R., and Mazzine, M. (2021). The synthesis of computational imprecision: Impact of compilation techniques on the approximate computing paradigm. ACM Computing Surveys (CSUR), 54(8):1–35.

Bharti, S., Kumar, A., and Gupta, P. (2022). Review of approximate computing in image processing applications. In 2022 4th International Conference on Artificial Intelligence and Speech Technology (AIST), pages 1–6.

Borisov, V., Leemann, T., Seßler, K., Haug, J., Pawelczyk, M., and Kasneci, G. (2024). Deep neural networks and tabular data: A survey. IEEE Transactions on Neural Networks and Learning Systems, 35(6):7499–7519.

Breiman, L. (2001). Random forests. Machine learning, 45(1):5–32.

Catelan, D. L. (2025). Approximate computing: Contributions to the design of arithmetic circuits and instruction-set architectures. Tese de doutorado, Universidade Federal de Mato Grosso do Sul (UFMS).

Chen, L., Han, J., Liu, W., and Lombardi, F. (2015). Design of approximate unsigned integer non-restoring divider for inexact computing. In Proceedings of the 25th edition on Great Lakes Symposium on VLSI, pages 51–56.

Das, A., Ghosh, S. K., Raha, A., and Raghunathan, V. (2024). Toward energy-efficient collaborative inference using multisystem approximations. IEEE Internet of Things Journal, 11(10):17989–18004.

Deb, K., Agrawal, S., Pratap, A., and Meyarivan, T. (2000). A fast elitist non-dominated sorting genetic algorithm for multi-objective optimization: NSGA-II. In International Conference on Parallel Problem Solving From Nature, pages 849–858. Springer.

Geurts, P., Ernst, D., and Wehenkel, L. (2006). Extremely randomized trees. Machine learning, 63(1):3–42.

Graham, S. L., Kessler, P. B., and Mckusick, M. K. (1982). Gprof: A call graph execution profiler. SIGPLAN Not., 17(6):120–126.

Hoffmann, H., Misailovic, S., Sidiroglou, S., Agarwal, A., and Rinard, M. (2009). Using code perforation to improve performance, reduce energy consumption, and respond to failures. Technical report, MIT Computer Science and Artificial Intelligence Laboratory (CSAIL), Massachusetts Institute of Technology.

Ke, G., Meng, Q., Finley, T., Wang, T., Chen, W., Ma, W., Ye, Q., and Liu, T.-Y. (2017). Lightgbm: A highly efficient gradient boosting decision tree. Advances in neural information processing systems, 30.

Leon, V., Hanif, M. A., Armeniakos, G., Jiao, X., Shafique, M., Pekmestzi, K., and Soudris, D. (2025). Approximate computing survey, part i: Terminology and software & hardware approximation techniques. ACM Computing Surveys, 57(7):1–36.

Misailovic, S., Carbin, M., Achour, S., Qi, Z., and Rinard, M. C. (2014). Chisel: Reliability-and accuracy-aware optimization of approximate computational kernels. ACM Sigplan Notices, 49(10):309–328.

Mittal, S. (2016). A survey of techniques for approximate computing. ACM Computing Surveys (CSUR), 48(4):1–33.

Mrazek, V., Hanif, M. A., Vasicek, Z., Sekanina, L., and Shafique, M. (2019). autoax: An automatic design space exploration and circuit building methodology utilizing libraries of approximate components. In Proceedings of the 56th Annual Design Automation Conference 2019, page 1–6. ACM.

Muchnick, S. S. (1997). Advanced Compiler Design and Implementation. Morgan Kaufmann.

Neto, G. K., Sovernigo, F., Catelan, D., Santos, R., and Duenha, L. (2025). Heuristics for energy-efficient instruction-level approximate computing. In 2025 IEEE/SBC 37th International Symposium on Computer Architecture and High Performance Computing (SBAC-PAD), pages 35–45. IEEE.

Rasheed, Y., Anwar, M. W., Gillani, G. A., and Ottavi, M. (2025). Toward approximate computing for deep learning in embedded systems: A systematic literature review. IEEE Access, 13:210863–210891.

Saeedi, S., Savino, A., and Di Carlo, S. (2023). Design space exploration of approximate computing techniques with a reinforcement learning approach. In 2023 53rd Annual IEEE/IFIP International Conference on Dependable Systems and Networks Workshops (DSN-W).

Sharif, H., Zhao, Y., Kotsifakou, M., Kothari, A., Schreiber, B., Wang, E., Sarita, Y., Zhao, N., Joshi, K., Adve, V. S., et al. (2021). Approxtuner: a compiler and runtime system for adaptive approximations. In Proceedings of the 26th ACM SIGPLAN Symposium on Principles and Practice of Parallel Programming, pages 262–277.

Yazdanbakhsh, A., Mahajan, D., Esmaeilzadeh, H., and Lotfi-Kamran, P. (2017). Axbench: A multiplatform benchmark suite for approximate computing. IEEE Design and Test, 34(2):60–68.