The use of perplexity as model selection criterion seems misguided to me. See my Quora answer to the question "What is the relationship between perplexity and Kolmogorov complexity? <https://www.quora.com/What-is-the-relationship-between-perplexity-and-Kolmogorov-complexity/answer/James-Bowery>" I'd say "antiquated" rather than "misguided" but AFAIK Solomonoff's papers on the use of Kolmogorov Complexity as universal model selection criterion predated perplexity's use as language model selection criterion.
------------------------------------------ Artificial General Intelligence List: AGI Permalink: https://agi.topicbox.com/groups/agi/Tbc419b4c00dd690d-M2db242ad0bb4621f933dfc3c Delivery options: https://agi.topicbox.com/groups/agi/subscription