HardCoRe-NAS: Hard Constrained diffeRentiable Neural Architecture Search
This addresses the need for efficient and reliable network design in resource-constrained applications, representing an incremental improvement over previous soft-constraint methods.
The paper tackles the problem of neural networks not strictly adhering to resource constraints like latency and memory in Neural Architecture Search (NAS) by introducing HardCoRe-NAS, which enforces hard constraints throughout the search, resulting in state-of-the-art architectures that exactly meet these constraints without tuning.
Realistic use of neural networks often requires adhering to multiple constraints on latency, energy and memory among others. A popular approach to find fitting networks is through constrained Neural Architecture Search (NAS), however, previous methods enforce the constraint only softly. Therefore, the resulting networks do not exactly adhere to the resource constraint and their accuracy is harmed. In this work we resolve this by introducing Hard Constrained diffeRentiable NAS (HardCoRe-NAS), that is based on an accurate formulation of the expected resource requirement and a scalable search method that satisfies the hard constraint throughout the search. Our experiments show that HardCoRe-NAS generates state-of-the-art architectures, surpassing other NAS methods, while strictly satisfying the hard resource constraints without any tuning required.