Fwd: Reminder - Thesis Proposal - 1/12/18 - Manzil Zaheer - Representation Learning @ Scale

[Artur Dubrawski]

if you're free please attend Manzil's presentation


-------- Forwarded Message --------
Subject: 	Reminder - Thesis Proposal - 1/12/18 - Manzil Zaheer - 
Representation Learning @ Scale
Date: 	Thu, 11 Jan 2018 17:09:00 -0500
From: 	Diane Stidle <diane+ at cs.cmu.edu>
To: 	ml-seminar at cs.cmu.edu <ML-SEMINAR at cs.cmu.edu>, Alex Smola 
<alex.smola at gmail.com>, mccallum at cs.umass.edu <mccallum at cs.umass.edu>



/Thesis Proposal/

Date: January 12, 2018
Time: 3:00 PM
Place: 8102 GHC
Speaker: Manzil Zaheer

Title: Representation Learning @ Scale

Abstract:

Machine learning techniques are reaching or exceeding human level 
performances in tasks like image classification, translation, and 
text-to-speech. The success of these machine learning algorithms have 
been attributed to highly versatile representations learnt from data 
using deep networks or intricately designed Bayesian models*. 
*_*Represen*_*tation learning* has also provided hints in neuroscience, 
e.g. for understanding how humans might categorize objects. Despite 
these instances of success, many open questions remain.

Data come in all shapes and sizes: not just as images or text, but also 
as point clouds, sets, graphs, compressed, or even heterogeneous mixture 
of these data types. In this thesis, we want to develop
representation learning algorithms for such unconventional data types by 
leveraging their structure and establishing new mathematical properties. 
Representations learned in this fashion were applied on diversedomains 
and found to be competitive with task specific state-of-the-art methods.

Once we have the representations, in various applications its 
interpretability is as crucial as its accuracy. Deep models often yield 
better accuracy, but require a large number of parameters, often 
notwithstanding the simplicity of the underlying data, rendering it 
uninterpretable which is highly undesirable in tasks
like user modeling. On the other hand, Bayesian models produce sparse 
discrete representations, easily amenable to human interpretation. In 
this thesis, we want to explore methods *that are**capable of *
*learning *mixed representations retaining best of both the worlds. Our 
experimental evaluations show that the proposed techniques compare 
favorably with several state-of-the-art baselines.

Finally, one would want such interpretable representations to be 
inferred from large-scale data, however, often there is a mismatch 
between our computational resources and the statistical models. In this 
thesis, we want to bridge this gap by solutions based on a combination 
of modern computational techniques/data structures on one side and 
modified statistical inference algorithms on the other. We introduce new 
ways to parallelize, reduce look-ups, handle variable state space size, 
and escape saddle points. On latent variable models, like latent 
Dirichlet allocation (LDA), we find significant gains in performance.

To summarize, in this thesis, we want to explore three major aspects of 
representation learning --- diversity: being able to handle different 
types of data, interpretability: being accessible to and understandable 
by humans, and scalablity: being able to process massive datasets in a 
reasonable time and budget.

Thesis Committee:
     Barnabas Poczos, Co-Chair
     Ruslan Salakhutdinov, Co-Chair
     Alexander J Smola (Amazon)
     Andrew McCallum (UMass Amherst)

Link to proposal document:
http://manzil.ml/proposal.pdf​

-- 
Diane Stidle
Graduate Programs Manager
Machine Learning Department
Carnegie Mellon University
diane at cs.cmu.edu
412-268-1299

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.srv.cs.cmu.edu/pipermail/autonlab-research/attachments/20180111/ffff3591/attachment.html>