aug 8 & 9

Author: pwyq

2023/index.html

@@ -197,14 +197,12 @@ <h1>2023 Tea Time Talks</h1>
       </p>
       <br>
 
-      <p>
-        <b>Samuel Robertson (August 8, 2023; CSC 3-33)</b><br>
-        <i>Computational and Statistical Bounds for Linear Bandits</i><br>
-      </p>
-      <br>
-
       <p>
         <b>Homayoon Farrahi (August 9, 2023; Amii)</b><br>
+        <i>Automatic Meta-Descent</i>
+        <p align="justify">
+          Online continual-learning agents need to adapt to the nonstationarities in the environment. Since the environment can combine stationary and nonstationary components, each weight of the learning agent can benefit from its own adjustable step size. In meta-descent, both the weight and step-size values are learned using gradient descent to optimize the objective of the problem. We show that the meta-step size of an existing meta-descent algorithm has to be tuned to perform reasonably on different problems. We propose a novel meta-descent algorithm using the normalization technique of Adam and empirically compare it with Adam in a continuous-control reinforcement learning task.
+        </p>
       </p>
       <br>