Roy Fox, Oct 2014 - Revision history

Ipapusha at 18:41, 29 September 2014

2014-09-29T18:41:52Z

← Older revision		Revision as of 18:41, 29 September 2014
Line 6:		Line 6:
	* 12:00 pm - Group meeting talk, 106 ANB		* 12:00 pm - Group meeting talk, 106 ANB
	* 1:30 pm - Ioannis Filippidis		* 1:30 pm - Ioannis Filippidis
	* 2:15 pm - ~~Open~~		* 2:15 pm - Ivan Papusha
	* 3:00 pm - Yilin Mo		* 3:00 pm - Yilin Mo
	* 3:45 pm - Open		* 3:45 pm - Open

Ifilippi: add Ioannis

2014-09-29T18:05:50Z

add Ioannis

← Older revision		Revision as of 18:05, 29 September 2014
Line 5:		Line 5:
	* 11:30 am - Richard Murray		* 11:30 am - Richard Murray
	* 12:00 pm - Group meeting talk, 106 ANB		* 12:00 pm - Group meeting talk, 106 ANB
	* 1:30 pm - ~~Open~~		* 1:30 pm - Ioannis Filippidis
	* 2:15 pm - Open		* 2:15 pm - Open
	* 3:00 pm - Yilin Mo		* 3:00 pm - Yilin Mo

Yilinmo: /* Schedule */

2014-09-29T16:42:41Z

Schedule

← Older revision		Revision as of 16:42, 29 September 2014
Line 7:		Line 7:
	* 1:30 pm - Open		* 1:30 pm - Open
	* 2:15 pm - Open		* 2:15 pm - Open
	* 3:00 pm - ~~Open~~		* 3:00 pm - Yilin Mo
	* 3:45 pm - Open		* 3:45 pm - Open
	* 4:30 pm - Depart		* 4:30 pm - Depart

Murray: Murray moved page Roy Fox, Sep 2014 to Roy Fox, Oct 2014

2014-09-26T20:22:13Z

Murray moved page Roy Fox, Sep 2014 to Roy Fox, Oct 2014

← Older revision	Revision as of 20:22, 26 September 2014
(No difference)

Murray: /* Abstract */

2014-09-26T20:16:53Z

Abstract

← Older revision		Revision as of 20:16, 26 September 2014
Line 19:		Line 19:
	106 Annenberg		106 Annenberg

	Intelligent agents, interacting with their environment, operate under constraints on what they can observe and how they can act. Unbounded agents can use standard Reinforcement Learning to optimize their inference and control under purely external constraints. Bounded agents, on the other hand, are subject to internal constraints as well. This only allows them to partially notice their observations, and to partially intend their actions, requiring rational selection of attention and action.		Intelligent agents, interacting with their environment, operate under constraints on what they can observe and how they can act. Unbounded agents can use standard Reinforcement Learning to optimize their inference and control under purely external constraints. Bounded agents, on the other hand, are subject to internal constraints as well. This only allows them to partially notice their observations, and to partially intend their actions, requiring rational selection of attention and action.

	In this talk we will see how to find the optimal information-constrained policy in reactive (memoryless) agents. We will discuss a number of reasons why internal constraints are often best modeled as bounds on information-theoretic quantities, and why we can focus on reactive agents with hardly any loss of generality. We will link the solution of the constrained problem to that of soft clustering, and present some of its nice properties, such as principled dimensionality reduction.		In this talk we will see how to find the optimal information-constrained policy in reactive (memoryless) agents. We will discuss a number of reasons why internal constraints are often best modeled as bounds on information-theoretic quantities, and why we can focus on reactive agents with hardly any loss of generality. We will link the solution of the constrained problem to that of soft clustering, and present some of its nice properties, such as principled dimensionality reduction.

Murray: Created page with "Roy Fox is a 4th year PhD student in the Computer Science working with Prof. Naftali Tishby on Information Theoretic methods for Reinforcement Learning. He will be visiting C..."

2014-09-26T20:16:35Z

Created page with "Roy Fox is a 4th year PhD student in the Computer Science working with Prof. Naftali Tishby on Information Theoretic methods for Reinforcement Learning. He will be visiting C..."

New page

Roy Fox is a 4th year PhD student in the Computer Science working with Prof. Naftali Tishby on Information Theoretic methods for Reinforcement Learning. He will be visiting Caltech on 2 Oct (Thu). If you would like to meet with him, please sign up for a slot below (use your IMSS credentials to log in).

=== Schedule ===

* 11:30 am - Richard Murray
* 12:00 pm - Group meeting talk, 106 ANB
* 1:30 pm - Open
* 2:15 pm - Open
* 3:00 pm - Open
* 3:45 pm - Open
* 4:30 pm - Depart

=== Abstract ===

Optimal Selective Attention and Action in Reactive Agents <br>
Roy Fox, Hebrew University

2 Oct (Thu), 12 pm <br>
106 Annenberg

Intelligent agents, interacting with their environment, operate under constraints on what they can observe and how they can act. Unbounded agents can use standard Reinforcement Learning to optimize their inference and control under purely external constraints. Bounded agents, on the other hand, are subject to internal constraints as well. This only allows them to partially notice their observations, and to partially intend their actions, requiring rational selection of attention and action.

In this talk we will see how to find the optimal information-constrained policy in reactive (memoryless) agents. We will discuss a number of reasons why internal constraints are often best modeled as bounds on information-theoretic quantities, and why we can focus on reactive agents with hardly any loss of generality. We will link the solution of the constrained problem to that of soft clustering, and present some of its nice properties, such as principled dimensionality reduction.