<html>
<head>
<meta http-equiv="Content-Type" content="text/html; charset=utf-8">
</head>
<body style="word-wrap: break-word; -webkit-nbsp-mode: space; -webkit-line-break: after-white-space;" class="">
<div class="">This is an announcement of Horace Pan's Dissertation Defense.</div>
<div class="">===============================================</div>
<div class="">Candidate: Horace Pan</div>
<div class=""><br class="">
</div>
<div class="">Date: Wednesday, June 01, 2022</div>
<div class=""><br class="">
</div>
<div class="">Time: 11 am CST</div>
<div class=""><br class="">
</div>
<div class="">Remote Location: <a href="https://uchicago.zoom.us/j/96060037177?pwd=VW44TUxablhsSk55b2dSdlEzbE45Zz09" class="">https://uchicago.zoom.us/j/96060037177?pwd=VW44TUxablhsSk55b2dSdlEzbE45Zz09</a> Meeting ID: 960 6003 7177 Passcode: 876833</div>
<div class=""><br class="">
</div>
<div class=""><br class="">
</div>
<div class="">Title: Leveraging Structure and Symmetry in Machine Learning</div>
<div class=""><br class="">
</div>
<div class="">Abstract: In this thesis we describe two separate works: higher order permutation equivariant layers for neural networks and Fourier bases for reinforcement learning over combinatorial puzzles.</div>
<div class=""><br class="">
</div>
<div class="">Recent work on permutation equivariant neural networks has mostly focused on the first order case (sets) and the second order case (graphs). We describe the machinery for generalizing permutation equivariance to arbitrary k-ary interactions and
provide a systematic approach for efficiently computing these kth-order permutation equivariant layer and enumerating all the intermediate operations involved. We evaluate our proposed permutation equivariant architectures using higher order equivariant layers
on a variety of set learning tasks and find that our models are competitive with existing baseline methods while using much fewer parameters.</div>
<div class=""><br class="">
</div>
<div class="">Traditionally, permutation puzzles such as the Rubik's Cube were often solved by heuristic search like A*-search and value based reinforcement learning methods. Both heuristic search and Q-learning approaches to solving these puzzles can be reduced
to learning a heuristic/value function to decide what puzzle move to make at each step. We propose learning a value function using the irreducible representations basis (which we will also call the Fourier basis) of the puzzle’s underlying group. Classical
Fourier analysis on real valued functions tells us we can approximate smooth functions with low frequency basis functions. Similarly, smooth functions on finite groups can be represented by the analogous low frequency Fourier basis functions. We demonstrate
that we can learn effective value functions in the Fourier basis for solving various permutation puzzles using fewer parameters and fewer samples than deep value networks.</div>
<div class=""><br class="">
</div>
<div class="">Advisors: Risi Kondor</div>
<div class=""><br class="">
</div>
<div class="">Committee Members: Risi Kondor, Yuxin Chen, and Eric Jonas</div>
<div class=""><br class="">
</div>
<div class=""><br class="">
</div>
<div class=""><span class="" style="font-family: Calibri, Arial, Helvetica, sans-serif; font-size: 16px;"> </span><a href="http://www.people.cs.uchicago.edu/~hopan/horacepan_thesis_draft.pdf" id="LPlnk519523" class="" style="font-family: Calibri, Arial, Helvetica, sans-serif; font-size: 16px;">http://www.people.cs.uchicago.edu/~hopan/horacepan_thesis_draft.pdf</a></div>
</body>
</html>