Title:
Better Intelligence If and Only If Better Compression
Better Intelligence If and Only If Better Compression
Speaker:
王富民 (Fumin Wang), PhD, DeepMind
王富民 (Fumin Wang), PhD, DeepMind
Time:
11/23 (Sat.) 5 pm PST, 6 pm MST, 7 pm CST, 8 pm EST
11/24 (Sun.) 9 am Taiwan
11/23 (Sat.) 5 pm PST, 6 pm MST, 7 pm CST, 8 pm EST
11/24 (Sun.) 9 am Taiwan
Keywords:
Machine learning, Information theory, Minimum description length principle, Rissanen bound, Kolmogorov complexity
Machine learning, Information theory, Minimum description length principle, Rissanen bound, Kolmogorov complexity
Abstract:
In the theory of Artificial General Intelligence by Hutter and others, the optimal behavior of a rational agent is equivalent to compressing its observations. A similar statement is Occam's Razor, i.e. the simplest answer is usually the correct answer. In this talk, we provide a compression program that is stronger than zip, 7z, and tar.gz, and show how it can be used to make intelligent decisions. In particular, we apply it to DNA and protein codes and use the resulting complexity measures to reconstruct the phylogeny tree of the mammalian class, as well as that of the SARS virus.
In the theory of Artificial General Intelligence by Hutter and others, the optimal behavior of a rational agent is equivalent to compressing its observations. A similar statement is Occam's Razor, i.e. the simplest answer is usually the correct answer. In this talk, we provide a compression program that is stronger than zip, 7z, and tar.gz, and show how it can be used to make intelligent decisions. In particular, we apply it to DNA and protein codes and use the resulting complexity measures to reconstruct the phylogeny tree of the mammalian class, as well as that of the SARS virus.
No comments:
Post a Comment