January 28, 2015<\/p>\n
<\/p>\n
Attending:<\/p>\n
Thom<\/p>\n
Bill<\/p>\n
Karthik<\/p>\n
Nic<\/p>\n
Lee<\/p>\n
Mike<\/p>\n
Wren<\/p>\n
Frode<\/p>\n
<\/p>\n
Agenda:<\/p>\n
Gecco Papers -3<\/p>\n
Thom, Bill, Nic<\/p>\n
<\/p>\n
Lee-<\/p>\n
Asymmetry on parents<\/p>\n
Collaboration with David Clark on math-finite terms and finite algebra, Clark has algorithms that are not evolutionary<\/p>\n
Pucks; bonding, will then be able to do push based<\/p>\n
<\/p>\n
Nic-<\/p>\n
Crossover revised paper; 2 problems, crossover bias and tournaments<\/p>\n
<\/p>\n
Lee and Nic-<\/p>\n
Measurement of success in systems, best approximation, best fitness<\/p>\n
Both metrics are potentially useful, and when you don\u2019t know\u2026<\/p>\n
<\/p>\n
Thom and Nic-<\/p>\n
Generalization or how well matches the training set-over fitting?<\/p>\n
No validation, but spot-check via graphs; stable, solid approximations over test area<\/p>\n
Adding crossover bias-fitness becomes larger<\/p>\n
<\/p>\n
Bill-<\/p>\n
Generalization and how to promote it in GP runs<\/p>\n
<\/p>\n
Lee, Thom and Bill-<\/p>\n
Papers from journal; collectively read and discuss<\/p>\n
Trade of fitness and validation of fitness<\/p>\n
How different system generalize-but not making better at generalization<\/p>\n
<\/p>\n
Bill-<\/p>\n
Promoting generalization in symbolic regression<\/p>\n
Smaller programs generally generalize better with better fitness<\/p>\n
Random mate selection, survival step-fitness and fitness generality<\/p>\n
<\/p>\n
Lee, Thom and Bill-<\/p>\n
2 fitness cases,<\/p>\n
Survival-how well fitness to 1, and how well fitness\u2019s compare<\/p>\n
2 sets similar-performance on one dependent on other<\/p>\n
Size connects to post run simplification<\/p>\n
Simplify with hill climber after<\/p>\n
Each step turns some genes off (remove sub-expression)<\/p>\n
<\/p>\n
Lee-<\/p>\n
Trees; mutations with replacement of sub tree with parent tree<\/p>\n
Preserve syntax<\/p>\n
<\/p>\n
Lee, Bill and Thom-<\/p>\n
Post run simplification<\/p>\n
Instead of pure hill climbing<\/p>\n
Turn off two genes and turn one gene back on<\/p>\n
Higher probability of turning things off then on<\/p>\n
Frode tried to do systematic testing; same simplification but as genetic operator and how it works with size and generalization; all a wash, simplifying during a run has all sorts of repercussions<\/p>\n
Post run simplification for generalization<\/p>\n
<\/p>\n
Bill, Nic and Lee-<\/p>\n
Epigenetic paper; danger of varying multiple things and their repercussions and relations within the system<\/p>\n
What is the epigenetics contributing? Is it just turning off your other additions?<\/p>\n
Tough sell without solving some unsolvable problems<\/p>\n
Clarity about base system and what is being changed<\/p>\n
Explanation of the nuances of ideas<\/p>\n
Rational for starting point<\/p>\n
2 stack based systems; epigenetics is easier to do in stack-based systems<\/p>\n
What happens when stuff gets turned off?<\/p>\n
<\/p>\n
Thom-<\/p>\n
Push for hill climbing; doesn\u2019t take secondary<\/p>\n
Keeps child if better at every test case, even with out lexicase<\/p>\n
<\/p>\n
Wren<\/p>\n
More LatB data<\/p>\n
Calc concentration eq; relates to what the cell looks like<\/p>\n
Models with Sarah<\/p>\n
<\/p>\n
Mike<\/p>\n
Documentation; example programs, how to put in your own data<\/p>\n
Java version of push<\/p>\n
Unwitting GP, Parasitic computing<\/p>\n
Zeke\u2019s tree based GP<\/p>\n
Quill, Gorilla repl<\/p>\n
Outsiders get started with out intensive tutorial<\/p>\n
Cleaner; FourPush?<\/p>\n
<\/p>\n
Karthik<\/span><\/strong><\/p>\n <\/p>\n Trying inductive synthesis with Sketch (having trouble with strings)<\/p>\n Running Tom\u2019s Benchmarks<\/p>\n <\/p>\n Frode<\/span><\/strong><\/p>\n Thinking about using simplification before parent selection.<\/p>\n Unclear whether this would help.<\/p>\n <\/p>\n Tom<\/span><\/strong><\/p>\n GECCO Papers<\/p>\n <\/p>\n Asymmetry<\/span><\/strong><\/p>\n -In tree based GP, the root node is more important. <\/p>\n *GPTP wants \u201csomething\u201d by 2\/7.<\/strong><\/p>\n *Gecco Workshop deadlines: 4\/3<\/strong><\/p>\n","protected":false},"excerpt":{"rendered":" January 28, 2015 Attending: Thom Bill Karthik Nic Lee Mike Wren Frode Agenda: Gecco Papers -3 Thom, Bill, Nic Lee- Asymmetry on parents Collaboration with David Clark on math-finite terms and finite algebra, Clark has algorithms that are not evolutionary Pucks; bonding, will then be able to do push based Nic- […]<\/p>\n","protected":false},"author":916,"featured_media":0,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[1],"tags":[],"class_list":["post-684","post","type-post","status-publish","format-standard","hentry","category-uncategorized"],"_links":{"self":[{"href":"https:\/\/sites.hampshire.edu\/ci-lab\/wp-json\/wp\/v2\/posts\/684"}],"collection":[{"href":"https:\/\/sites.hampshire.edu\/ci-lab\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/sites.hampshire.edu\/ci-lab\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/sites.hampshire.edu\/ci-lab\/wp-json\/wp\/v2\/users\/916"}],"replies":[{"embeddable":true,"href":"https:\/\/sites.hampshire.edu\/ci-lab\/wp-json\/wp\/v2\/comments?post=684"}],"version-history":[{"count":1,"href":"https:\/\/sites.hampshire.edu\/ci-lab\/wp-json\/wp\/v2\/posts\/684\/revisions"}],"predecessor-version":[{"id":685,"href":"https:\/\/sites.hampshire.edu\/ci-lab\/wp-json\/wp\/v2\/posts\/684\/revisions\/685"}],"wp:attachment":[{"href":"https:\/\/sites.hampshire.edu\/ci-lab\/wp-json\/wp\/v2\/media?parent=684"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/sites.hampshire.edu\/ci-lab\/wp-json\/wp\/v2\/categories?post=684"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/sites.hampshire.edu\/ci-lab\/wp-json\/wp\/v2\/tags?post=684"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}
\n-Biological Reproduction is generally asymmetric<\/p>\n