[Colloquium] REMINDER: 3/14 Talks at TTIC: Nan Jiang, University of Michigan

[Mary Marre via Colloquium]

When:     Tuesday, March 14th at 11:00 am

Where:    TTIC, 6045 S Kenwood Avenue, 5th Floor, Room 526

Who:       Nan Jiang, University of Michigan


Title:        New Results in Statistical Reinforcement Learning

Abstract:
Recently, reinforcement learning (RL) has achieved inspiring success in game
playing domains, including human-level control in Atari games and mastering
the game of Go. Looking into the future, we expect to build machine learning
 systems that use RL to turn predictions into actions; applications include
robotics, dialog systems, online education, adaptive medical treatment, to
name but a few.

In this talk, I show how theoretical insights from supervised learning can
help understand RL, and better appreciate the unique challenges that arise
from multi-stage decision making. The first part of the talk focuses on an
interesting phenomenon, that a short planning horizon can produce better
policies when there is limited data. I explain it by making a formal
analogy to empirical risk minimization, and argue that a short planning
horizon helps avoid overfitting. The second part of the talk concerns a
core algorithmic challenge in state-of-the-art RL: sample-efficient
exploration in large state spaces. I introduce a new complexity measure,
the Bellman rank, which allows us to apply a unified algorithm to a number
of important RL settings, in some cases obtaining polynomial sample
complexity for the first time.

Bio:
Nan Jiang is a PhD candidate in Computer Science and Engineering at
University of Michigan, He works with Satinder Singh on a variety of topics
related to reinforcement learning. Specific research interests include
provable use of function approximation, off-policy evaluation, state
representation learning, spectral learning of dynamical systems, and
inverse RL for AI safety. Nan received his bachelor degree in Control and
Automation from Tsinghua University in 2011. He received the Best Paper
Award at AAMAS 2015, and Rackham Predoctoral Fellowship in 2016.


Host: Matthew Walter <mwalter at ttic.edu>






Mary C. Marre
Administrative Assistant
*Toyota Technological Institute*
*6045 S. Kenwood Avenue*
*Room 504*
*Chicago, IL  60637*
*p:(773) 834-1757*
*f: (773) 357-6970*
*mmarre at ttic.edu <mmarre at ttic.edu>*

On Mon, Mar 13, 2017 at 2:24 PM, Mary Marre <mmarre at ttic.edu> wrote:

> When:     Tuesday, March 14th at 11:00 am
>
> Where:    TTIC, 6045 S Kenwood Avenue, 5th Floor, Room 526
>
> Who:       Nan Jiang, University of Michigan
>
>
> Title:        New Results in Statistical Reinforcement Learning
>
> Abstract:
> Recently, reinforcement learning (RL) has achieved inspiring success in game
> playing domains, including human-level control in Atari games and
> mastering the game of Go. Looking into the future, we expect to build
> machine learning systems that use RL to turn predictions into actions;
> applications include robotics, dialog systems, online education, adaptive
> medical treatment, to name but a few.
>
> In this talk, I show how theoretical insights from supervised learning can
> help understand RL, and better appreciate the unique challenges that arise
> from multi-stage decision making. The first part of the talk focuses on an
> interesting phenomenon, that a short planning horizon can produce better
> policies when there is limited data. I explain it by making a formal
> analogy to empirical risk minimization, and argue that a short planning
> horizon helps avoid overfitting. The second part of the talk concerns a
> core algorithmic challenge in state-of-the-art RL: sample-efficient
> exploration in large state spaces. I introduce a new complexity measure,
> the Bellman rank, which allows us to apply a unified algorithm to a number
> of important RL settings, in some cases obtaining polynomial sample
> complexity for the first time.
>
> Bio:
> Nan Jiang is a PhD candidate in Computer Science and Engineering at
> University of Michigan, He works with Satinder Singh on a variety of topics
> related to reinforcement learning. Specific research interests include
> provable use of function approximation, off-policy evaluation, state
> representation learning, spectral learning of dynamical systems, and
> inverse RL for AI safety. Nan received his bachelor degree in Control and
> Automation from Tsinghua University in 2011. He received the Best Paper
> Award at AAMAS 2015, and Rackham Predoctoral Fellowship in 2016.
>
>
> Host: Matthew Walter <mwalter at ttic.edu>
>
>
>
>
>
> Mary C. Marre
> Administrative Assistant
> *Toyota Technological Institute*
> *6045 S. Kenwood Avenue*
> *Room 504*
> *Chicago, IL  60637*
> *p:(773) 834-1757 <(773)%20834-1757>*
> *f: (773) 357-6970 <(773)%20357-6970>*
> *mmarre at ttic.edu <mmarre at ttic.edu>*
>
> On Wed, Mar 8, 2017 at 10:05 AM, Mary Marre <mmarre at ttic.edu> wrote:
>
>> When:     Tuesday, March 14th at 11:00 am
>>
>> Where:    TTIC, 6045 S Kenwood Avenue, 5th Floor, Room 526
>>
>> Who:       Nan Jiang, University of Michigan
>>
>>
>> Title:        New Results in Statistical Reinforcement Learning
>>
>> Abstract:
>> Recently, reinforcement learning (RL) has achieved inspiring success in game
>> playing domains, including human-level control in Atari games and
>> mastering the game of Go. Looking into the future, we expect to build
>> machine learning systems that use RL to turn predictions into actions;
>> applications include robotics, dialog systems, online education, adaptive
>> medical treatment, to name but a few.
>>
>> In this talk, I show how theoretical insights from supervised learning can
>> help understand RL, and better appreciate the unique challenges that arise
>> from multi-stage decision making. The first part of the talk focuses on an
>> interesting phenomenon, that a short planning horizon can produce better
>> policies when there is limited data. I explain it by making a formal
>> analogy to empirical risk minimization, and argue that a short planning
>> horizon helps avoid overfitting. The second part of the talk concerns a
>> core algorithmic challenge in state-of-the-art RL: sample-efficient
>> exploration in large state spaces. I introduce a new complexity measure,
>> the Bellman rank, which allows us to apply a unified algorithm to a number
>> of important RL settings, in some cases obtaining polynomial sample
>> complexity for the first time.
>>
>> Bio
>> Nan Jiang is a PhD candidate in Computer Science and Engineering at
>> University of Michigan, He works with Satinder Singh on a variety of topics
>> related to reinforcement learning. Specific research interests include
>> provable use of function approximation, off-policy evaluation, state
>> representation learning, spectral learning of dynamical systems, and
>> inverse RL for AI safety. Nan received his bachelor degree in Control
>> and Automation from Tsinghua University in 2011. He received the Best
>> Paper Award at AAMAS 2015, and Rackham Predoctoral Fellowship in 2016.
>>
>>
>> Host: Matthew Walter <mwalter at ttic.edu>
>>
>>
>>
>> Mary C. Marre
>> Administrative Assistant
>> *Toyota Technological Institute*
>> *6045 S. Kenwood Avenue*
>> *Room 504*
>> *Chicago, IL  60637*
>> *p:(773) 834-1757 <(773)%20834-1757>*
>> *f: (773) 357-6970 <(773)%20357-6970>*
>> *mmarre at ttic.edu <mmarre at ttic.edu>*
>>
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.cs.uchicago.edu/pipermail/colloquium/attachments/20170314/eeec5b84/attachment-0001.html>