|COOKIES: By using this website you agree that we can place Google Analytics Cookies on your device for performance monitoring.|
Bundle methods and its application in machine learning
If you have a question about this talk, please contact Shakir Mohamed.
We will present work on bundle methods for machine learning. In this talk, we would like to first briefly review some basic concepts in convex optimization, using quadratic programming as an example. After that, we will give some examples of convex objective functions that are widely used in machine learning. Then we will talk about the cutting-plane method and bundle methods, together with the convergence analysis. We may also talk a little bit on how these methods can be extended to the optimization of non-convex functions.
The following two papers can be used as our references:
Trinh-Minh-Tri Do and Thierry Arti`eres, “Large Margin Training for Hidden Markov Models with Partially Observed States”, in Proc. ICML -2009.
Alexander, J.S. ,Vishwanathan, S.V.N. and Quoc V.L. “Bundle methods for machine learning”, in Proc.NIPS-2007
This talk is part of the Machine Learning Reading Group @ CUED series.
This talk is included in these lists:
Note that ex-directory lists are not shown.
Other listsICT4D: ICT for Development Centre of International Studies Lecture Series Dimer observables and Cauchy-Riemann operators
Other talksSome Network Analysis Problems Motivated by the Smart Grid Conserving manuscripts for COLOUR FEniCS-Shells: solving thin structural theories using a high-level finite element language Race, Representation and Visibility Superconducting quantum circuits: qubits and not only (Prof. Yuri Pashkin, Lancaster University) Mapping grief: a conceptual framework for understanding the diverse spatialities of death, mourning and remembrance