Today’s issue of WorkCompRecap features a new Guest Post from Tom Warden, Chief Insurance and Science Officer at CLARA Analytics on Deep Learning technologies, and how neural networks, which had been somewhat constrained in the past by computing power and data availability, are now able to be constructed with exponentially more layers and nodes, enabling the networks to refine their modeling of a problem and create more accurate solutions.
Tom also describes some suitable applications for deep learning, outlines some necessary requirements, and provides some tips for getting deep learning sold to management, including narrowing the scope of efforts to conduct experiments and produce results quickly, rather than just viewing it as a magic wand to solve all problems.