Our next MSAIL discussion will focus on Stochastic Sorting, the art of sorting given a noisy comparator. We'll meet at: 3427 EECS, Wednesday, 2016-11-02, 18:00-19:00.So that we can buy an appropriate amount of snacks, please rsvp on Facebook. And bring all potentially interested friends! Sneakpeek of IdeasWe're familiar with MergeSort, HeapSort, QuickSort, and their ilk: they are comparison sorts that attain the provably best worstcase asymptotic complexity of O(n logn) comparisons.
But... what if we can't fully trust our comparator? What if there's noise? An algorithm robust to noisy comparators must solve that fundamental difficulty of Machine Learning: to synthesize uncertain observations into a more certain whole. The study of such algorithms, then, is part of Machine Learning, and in its connections to optimization, it relates not only to ML but also economics and even evolution. Check out the following for more details!Active Learning (Guillory et al 2009)Active Learning for Stochastic Sorting (Ailon 2012)
Sorting for Genetic Algorithms (Brownlee 2010) --MSAIL
But... what if we can't fully trust our comparator? What if there's noise? An algorithm robust to noisy comparators must solve that fundamental difficulty of Machine Learning: to synthesize uncertain observations into a more certain whole. The study of such algorithms, then, is part of Machine Learning, and in its connections to optimization, it relates not only to ML but also economics and even evolution. Check out the following for more details!Active Learning (Guillory et al 2009)Active Learning for Stochastic Sorting (Ailon 2012)
Sorting for Genetic Algorithms (Brownlee 2010) --MSAIL
Related Links
Co-Sponsored By
- Student Organization: Michigan Student Artificial Intelligence Lab
Explore Similar Events
-
Loading Similar Events...