The Power of Training: How Different Neural Network Setups Influence the Energy Demand

• URL: http://arxiv.org/abs/2401.01851v4
• Date: Sat, 05 Oct 2024 06:13:26 GMT
• Title: The Power of Training: How Different Neural Network Setups Influence the Energy Demand
• Authors: Daniel Geißler, Bo Zhou, Mengxi Liu, Sungho Suh, Paul Lukowicz,
• Abstract summary: This work offers a evaluation of the effects of variations in machine learning training regimes and learning paradigms on the energy consumption of computing, especially HPC hardware with a life-cycle aware perspective.
• Score: 5.526611783155303
• License:
• Abstract: This work offers a heuristic evaluation of the effects of variations in machine learning training regimes and learning paradigms on the energy consumption of computing, especially HPC hardware with a life-cycle aware perspective. While increasing data availability and innovation in high-performance hardware fuels the training of sophisticated models, it also fosters the fading perception of energy consumption and carbon emission. Therefore, the goal of this work is to raise awareness about the energy impact of general training parameters and processes, from learning rate over batch size to knowledge transfer. Multiple setups with different hyperparameter configurations are evaluated on three different hardware systems. Among many results, we have found out that even with the same model and hardware to reach the same accuracy, improperly set training hyperparameters consume up to 5 times the energy of the optimal setup. We also extensively examined the energy-saving benefits of learning paradigms including recycling knowledge through pretraining and sharing knowledge through multitask training.

Related papers

• From Computation to Consumption: Exploring the Compute-Energy Link for Training and Testing Neural Networks for SED Systems [9.658615045493734]
  We study several neural network architectures that are key components of sound event detection systems. We measure the energy consumption for training and testing small to large architectures. We establish complex relationships between the energy consumption, the number of floating-point operations, the number of parameters, and the GPU/memory utilization.
  arXiv  Detail & Related papers  (2024-09-08T12:51:34Z)
• Revisiting DNN Training for Intermittently Powered Energy Harvesting Micro Computers [0.6721767679705013]
  This study introduces and evaluates a novel training methodology tailored for Deep Neural Networks in energy-constrained environments. We propose a dynamic dropout technique that adapts to both the architecture of the device and the variability in energy availability. Preliminary results demonstrate that this strategy provides 6 to 22 percent accuracy improvements compared to the state of the art with less than 5 percent additional compute.
  arXiv  Detail & Related papers  (2024-08-25T01:13:00Z)
• Learning Iterative Reasoning through Energy Diffusion [90.24765095498392]
  We introduce iterative reasoning through energy diffusion (IRED), a novel framework for learning to reason for a variety of tasks. IRED learns energy functions to represent the constraints between input conditions and desired outputs. We show IRED outperforms existing methods in continuous-space reasoning, discrete-space reasoning, and planning tasks.
  arXiv  Detail & Related papers  (2024-06-17T03:36:47Z)
• Learning Lightweight Object Detectors via Multi-Teacher Progressive Distillation [56.053397775016755]
  We propose a sequential approach to knowledge distillation that progressively transfers the knowledge of a set of teacher detectors to a given lightweight student. To the best of our knowledge, we are the first to successfully distill knowledge from Transformer-based teacher detectors to convolution-based students.
  arXiv  Detail & Related papers  (2023-08-17T17:17:08Z)
• Experts Weights Averaging: A New General Training Scheme for Vision Transformers [57.62386892571636]
  We propose a training scheme for Vision Transformers (ViTs) that achieves performance improvement without increasing inference cost. During training, we replace some Feed-Forward Networks (FFNs) of the ViT with specially designed, more efficient MoEs. After training, we convert each MoE into an FFN by averaging the experts, transforming the model back into original ViT for inference.
  arXiv  Detail & Related papers  (2023-08-11T12:05:12Z)
• Minimizing Energy Consumption of Deep Learning Models by Energy-Aware Training [26.438415753870917]
  We propose EAT, a gradient-based algorithm that aims to reduce energy consumption during model training. We demonstrate that our energy-aware training algorithm EAT is able to train networks with a better trade-off between classification performance and energy efficiency.
  arXiv  Detail & Related papers  (2023-07-01T15:44:01Z)
• On Feature Diversity in Energy-based Models [98.78384185493624]
  An energy-based model (EBM) is typically formed of inner-model(s) that learn a combination of the different features to generate an energy mapping for each input configuration. We extend the probably approximately correct (PAC) theory of EBMs and analyze the effect of redundancy reduction on the performance of EBMs.
  arXiv  Detail & Related papers  (2023-06-02T12:30:42Z)
• Energy Efficiency of Training Neural Network Architectures: An Empirical Study [11.325530936177493]
  The evaluation of Deep Learning models has traditionally focused on criteria such as accuracy, F1 score, and related measures. The computations needed to train such models entail a large carbon footprint. We study the relations between DL model architectures and their environmental impact in terms of energy consumed and CO$$ emissions produced during training.
  arXiv  Detail & Related papers  (2023-02-02T09:20:54Z)
• Benchmarking Resource Usage for Efficient Distributed Deep Learning [10.869092085691687]
  We conduct over 3,400 experiments training an array of deep networks representing various domains/tasks. We fit power law models that describe how training time scales with available compute resources and energy constraints.
  arXiv  Detail & Related papers  (2022-01-28T21:24:15Z)
• Powerpropagation: A sparsity inducing weight reparameterisation [65.85142037667065]
  We introduce Powerpropagation, a new weight- parameterisation for neural networks that leads to inherently sparse models. Models trained in this manner exhibit similar performance, but have a distribution with markedly higher density at zero, allowing more parameters to be pruned safely. Here, we combine Powerpropagation with a traditional weight-pruning technique as well as recent state-of-the-art sparse-to-sparse algorithms, showing superior performance on the ImageNet benchmark.
  arXiv  Detail & Related papers  (2021-10-01T10:03:57Z)
• Towards the Systematic Reporting of the Energy and Carbon Footprints of Machine Learning [68.37641996188133]
  We introduce a framework for tracking realtime energy consumption and carbon emissions. We create a leaderboard for energy efficient reinforcement learning algorithms. We propose strategies for mitigation of carbon emissions and reduction of energy consumption.
  arXiv  Detail & Related papers  (2020-01-31T05:12:59Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.