NASBench-201 consists of 15,625 neural networks trained on the CIFAR-10, CIFAR-100 and ImageNet-16-120 datasets. Neural Networks in Network Design Spaces (NDS) uses the DARTS Liu et al. [2019] The networks are comprised of cells sampled from each of AmoebaNet Real et al. [2019], There exists approximately 5000 neural network architectures in each NDS design space. Our initial attempts at discovering ZC-NASMs took a different approach to program representation. To store intermediate scalars generated by the program, we allocate 20 memory addresses. ID and the third and fourth integers provide the read addresses for the operation.

Additionally, existing zero-cost proxies fail to generalize across neural architecture design spaces.

Neural Architecture Search (NAS) has significantly improved productivity in the design and deployment of neural networks (NN). As NAS typically evaluates multiple models by training them partially or completely, the improved productivity comes at the cost of significant carbon footprint. To alleviate this expensive training routine, zero-shot/cost proxies analyze an NN at initialization to generate a score, which correlates highly with its true accuracy. Zero-cost proxies are currently designed by experts conducting multiple cycles of empirical testing on possible algorithms, datasets, and neural architecture design spaces. This experimentation lowers productivity and is an unsustainable approach towards zero-cost proxy design as deep learning use-cases diversify in nature. Additionally, existing zero-cost proxies fail to generalize across neural architecture design spaces. In this paper, we propose a genetic programming framework to automate the discovery of zero-cost proxies for neural architecture scoring. Our methodology efficiently discovers an interpretable and generalizable zero-cost proxy that gives state of the art score-accuracy correlation on all datasets and search spaces of NASBench-201 and Network Design Spaces (NDS). We believe that this research indicates a promising direction towards automatically discovering zero-cost proxies that can work across network architecture design spaces, datasets, and tasks.