# Impact of ML Optimization Tactics on Greener Pre-Trained ML Models

arXiv: https://arxiv.org/abs/2409.12878 · HTML: https://mlsanity.com/paper/2409.12878/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- Builds on 19 of its 81 references that are in the map.
- First posted on arXiv 19 Sep 2024.

## Builds on (most cited first, 19 of 19)

- [Distilling the Knowledge in a Neural Network](https://mlsanity.com/paper/1503.02531/) (2015)
- [Fashion-MNIST: a Novel Image Dataset for Benchmarking Machine Learning Algorithms](https://mlsanity.com/paper/1708.07747/) (2017)
- [Winoground: Probing Vision and Language Models for Visio-Linguistic Compositionality](https://mlsanity.com/paper/2204.03162/) (2022)
- [The State of Sparsity in Deep Neural Networks](https://mlsanity.com/paper/1902.09574/) (2019)
- [Power Hungry Processing: Watts Driving the Cost of AI Deployment?](https://mlsanity.com/paper/2311.16863/) (2023)
- [The Carbon Footprint of Machine Learning Training Will Plateau, Then Shrink](https://mlsanity.com/paper/2204.05149/) (2022)
- [M3IT: A Large-Scale Dataset towards Multi-Modal Multilingual Instruction Tuning](https://mlsanity.com/paper/2306.04387/) (2023)
- [A Systematic Review of Green AI](https://mlsanity.com/paper/2301.11047/) (2023)
- [Software Engineering for AI-Based Systems: A Survey](https://mlsanity.com/paper/2105.01984/) (2021)
- [A Comprehensive Survey on Model Quantization for Deep Neural Networks in Image Classification](https://mlsanity.com/paper/2205.07877/) (2022)
- [A Survey of Methods for Low-Power Deep Learning and Computer Vision](https://mlsanity.com/paper/2003.11066/) (2020)
- [Artificial Intelligence Index Report 2024](https://mlsanity.com/paper/2405.19522/) (2024)
- [A Synthesis of Green Architectural Tactics for ML-Enabled Systems](https://mlsanity.com/paper/2312.09610/) (2023)
- [Energy Efficiency of Training Neural Network Architectures: An Empirical Study](https://mlsanity.com/paper/2302.00967/) (2023)
- [Lessons Learned from Mining the Hugging Face Repository](https://mlsanity.com/paper/2402.07323/) (2024)
- [EnergiBridge: Empowering Software Sustainability through Cross-Platform Energy Measurement](https://mlsanity.com/paper/2312.13897/) (2023)
- [GAISSALabel: A tool for energy labeling of ML models](https://mlsanity.com/paper/2401.17150/) (2024)
- [Innovating for Tomorrow: The Convergence of SE and Green AI](https://mlsanity.com/paper/2406.18142/) (2024)
- [Structured Model Pruning for Efficient Inference in Computational Pathology](https://mlsanity.com/paper/2404.08831/) (2024)

## Built on by (most cited first, 2 of 2)

- [How do Machine Learning Models Change?](https://mlsanity.com/paper/2411.09645/) (2024)
- [Aggregating empirical evidence from data strategy studies: a case on model quantization](https://mlsanity.com/paper/2505.00816/) (2025)

## Abstract (arXiv)

Background: Given the fast-paced nature of today's technology, which has surpassed human performance in tasks like image classification, visual reasoning, and English understanding, assessing the impact of Machine Learning (ML) on energy consumption is crucial. Traditionally, ML projects have prioritized accuracy over energy, creating a gap in energy consumption during model inference. Aims: This study aims to (i) analyze image classification datasets and pre-trained models, (ii) improve inference efficiency by comparing optimized and non-optimized models, and (iii) assess the economic impact of the optimizations. Method: We conduct a controlled experiment to evaluate the impact of various PyTorch optimization techniques (dynamic quantization, torch.compile, local pruning, and global pruning) to 42 Hugging Face models for image classification. The metrics examined include GPU utilization, power and energy consumption, accuracy, time, computational complexity, and economic costs. The models are repeatedly evaluated to quantify the effects of these software engineering tactics. Results: Dynamic quantization demonstrates significant reductions in inference time and energy consumption, making it highly suitable for large-scale systems. Additionally, torch.compile balances accuracy and energy. In contrast, local pruning shows no positive impact on performance, and global pruning's longer optimization times significantly impact costs. Conclusions: This study highlights the role of software engineering tactics in achieving greener ML models, offering guidelines for practitioners to make informed decisions on optimization methods that align with sustainability goals.