Democratizing the optimization of AI’s arcane neural networks – InfoWorld

Were only a few weeks into the new year, but already were seeing signs that automated machine learning modeling, sometimes known asautoML, is rising to a new plateau of sophistication.

Specifically, it appears that a promising autoML approach known as neural architecture search will soon become part of data scientists core toolkits. This refers to tools and methodologies for automating creation of optimized architectures for convolutional, recurrent, and other neural network architectures at the heart of AIs machine learning models.

Neural architecture search tools optimize the structure, weights, and hyperparameters of a machine learning models algorithmic neurons in order to make them more accurate, speedy, and efficient in performing data-driven inferences. This technology has only recently begun to emerge from labs devoted to basic research in AI tools and techniques. Theresearch literatureshows that neural architecture search tools have already outperformed manually designed neural nets in many AI R&D projects.

Go here to read the rest:
Democratizing the optimization of AI's arcane neural networks - InfoWorld