Alon Cohen - Between Online Learning and Reinforcement Learning

×

Error message

  • Deprecated function: Creation of dynamic property LdapUserConf::$createLDAPAccounts is deprecated in LdapUserConf->load() (line 265 of /var/lib/drupal7/modules/ldap/ldap_user/LdapUserConf.class.php).
  • Deprecated function: Creation of dynamic property LdapUserConf::$createLDAPAccountsAdminApproval is deprecated in LdapUserConf->load() (line 266 of /var/lib/drupal7/modules/ldap/ldap_user/LdapUserConf.class.php).

In this talk I will describe some of my work around Online Learning and Reinforcement Learning.
Online Learning is a classic sub-domain of Machine Learning that has provided endless contributions to fields such as Statistical Learning, Optimization, Decision Making and others.
Unlike Reinforcement Learning which focuses on planning long-term decisions in the face of a non-adversarial environment, Online Learning focuses on making short-term decisions in the face of an adversary - and doing so efficiently.
In my research, therefore, I am interested in the potential interface between these fields, looking to design efficient algorithms for long-term decision making in the face of a possibly adversarial environment---a problem with many real-life use cases.
I will focus on two works that leverage efficient Online Learning algorithms for learning Linear Quadratic Regulators and Stochastic Shortest Path problems.

Date and Time: 
Thursday, January 14, 2021 - 13:30 to 14:30
Speaker: 
Alon Cohen
Location: 
Zoom
Speaker Bio: 

Dr. Alon Cohen is a researcher at the Foundations of Machine Learning team in Google Tel-aviv, hosted by Prof. Yishay Mansour.
Dr. Cohen has published broadly in top-tier machine learning conferences such as ICML, NeuRIPS and COLT.
He has received a PhD in 2018 from the Faculty of IE&M at the Technion, where his advisor was Prof. Tamir Hazan.