University of Cambridge > Talks.cam > Inference Group > Information and Intelligence

  Browse A-Z

Log in

Cambridge users (raven)details Other usersdetails No account?details

Information on

Subscribing to talksdetails Finding a talkdetails Adding a talkdetails Disseminating talksdetails Help and Documentationdetails

Information and Intelligence

Add to your list(s) Download to your calendar using vCal

If you have a question about this talk, please contact Phil Cowans.

If we were to try to construct an intelligent machine, what, in very broad terms, would we be aiming to do? It can be argued that the answer is (roughly speaking) to compress data as efficiently as possible. This leads to the idea of “Kolmogorov Complexity” (and related measures), which also arises as the universal prior for probabilistic inference.

For fun, consider the following question: Suppose someone said that in one hour’s time you could have 1 second of computing time on a computer which executed 10 instructions per second, had a similarly large amount of memory, and started with a copy of the internet and all the libraries of the world (and possibly some other real world data) in its memory. At the end of the 1 second you get to keep a reasonable amount of output (say 1012 bytes). What program would you write? What would your answer be if the computer started off with a blank memory?

Some background reading for those interested:

http://www.idsia.ch/juergen/loconet/nngen.html http://www.geocities.com/jim_bowery/cprize.html http://people.cs.uchicago.edu/fortnow/papers/soph.pdf http://citeseer.ist.psu.edu/gacs93lecture.html http://homepages.cwi.nl/~paulv/papers/algorithmicstatistics.pdf

This talk is part of the Inference Group series.

Tell a friend about this talk:

This talk is included in these lists:

Note that ex-directory lists are not shown.

Other lists

Cambridge Area Sequencing Informatics Meeting VI (2014) Applied and Computational Analysis EDC Seminars

Other talks

Towards bulk extension of near-horizon geometries Intrinsically Motivating Teachers;STIR's use of Data Driven Insight to Iterate, Pivot and (where necessary) Fail Fast Sacred Mountains as Flood Refuge Sites in Northwest North America Production Processes Group Seminar - "Advanced water filtration platforms based on hierarchically structured carbon nanotubes." Development of a Broadly-Neutralising Vaccine against Blood-Stage P. falciparum Malaria The formation of high density dust rings and clumps: the role of vorticity A polyfold lab report Cyclic Peptides: Building Blocks for Supramolecular Designs Singularities of Hermitian-Yang-Mills connections and the Harder-Narasimhan-Seshadri filtration EU LIFE Lecture - "Histone Chaperones Maintain Cell Fates and Antagonize Reprogramming in C. elegans and Human Cells" The Partition of India and Migration Adaptation in log-concave density estimation
 

© 2006-2024 Talks.cam, University of Cambridge. Contact Us | Help and Documentation | Privacy and Publicity