Efficient Activation Function Optimization through Surrogate Modeling — mentioned by binghamgarrett

Thu, 07 Dec 2023 20:53:05 GMT

AQuaSurF is a new algorithm that automatically discovers better activation functions for your machine learning task, helping to improve performance. Here's how it works: • Embrace data-driven insights: AQuaSurF analyzes thousands of activation functions across diverse architectures and datasets, identifying what truly matters for performance. No more intuition-based guesswork! • Unleash the power of the AQuaSurF algorithm: This efficient algorithm leverages a unique surrogate performance measure to efficiently navigate the vast space of activation functions, uncovering hidden gems quickly. •…

Paper and discussion history. A mention is not an endorsement.

Efficient Activation Function Optimization through Surrogate Modeling — HN submission

Thu, 07 Dec 2023 20:53:04 GMT

AQuaSurF: Discover better activation functions for your ML task

2 points; 1 discussion reply on HN. Paper and source history. A submission is not an endorsement.

Efficient Activation Function Optimization through Surrogate Modeling · HackerPapers discussions

Efficient Activation Function Optimization through Surrogate Modeling — mentioned by binghamgarrett

Efficient Activation Function Optimization through Surrogate Modeling — HN submission