AI comes with a significant carbon footprint. Deep learning models’ training and inference require vast amounts of energy. The larger (and typically more predictively accurate) a model is, the more compute power it requires to run inference.
The core architectures of deep learning models can be optimized to increase performance efficiency and reduce models size without detracting from their accuracy. Smaller and more efficient models which require less processing power are both cheaper to run and much more friendly for the environment.
In this case study, discover how teams are building DNN architectures that consume anywhere from 30% to 80% less compute power for inference with Deci’s Neural Architecture Search (NAS) based deep learning development platform.
Complete the form to get immediate access to the case study.