Description: This article discussed the NEAT family of neural network algorithms. NEAT automatically determines the number of hidden neurons, as well as their structure. Additionally HyperNEAT can scale to very large problem domains.

Description: This article delves into a new line of statistical thinking of the author, that may be useful for product development actuaries. The key concept is how product design can produce products that diversify quickly, meaning that the company can be confident of a profitable product line, based on fewer sales than might be true for other designs.

Description: Presented at October 1985 Annual Meeting. Discusses: Measures which should be used in setting objectives and in evaluating performance for insurance companies and product lines, Variations that are appropriate for stock and mutual companies and United States and Canadian companies, Qualitative considerations in interpreting and evaluating information by the management reporting system.

Description: Read the Sept. 2017 issue of Small Talk published by Smaller Insurance Company Section. Practical information on trends that impact sections and activities.