Inductive Biases for Higher-Level Cognition Deep Learning

Most machine learning models employ the inductive bias, i.e., that all processes interact. This can lead to poor generalization (if data is limited) and lack of robustness to changing task distributions. Anirudh Goyal will talk about inductive biases to factorize knowledge into independent pieces so that they can be combined dynamically and can lead to systematic generalization.

📌 JOIN US ONLINE ZOOM
Please Join Zoom Meeting:
https://us02web.zoom.us/j/82206637658?pwd=ZE92a014VDdEOWsvR0x1UE9NRldtUT09

Meeting ID: 822 0663 7658
Passcode: 794352

📌 SCHEDULE
● 10:00 am Opening Remarks
● 10:05 am - 11:05 am Anirudh Goyal’s talk on Inductive Biases for Higher-Level Cognition Deep Learning
● 11:05 am - 11:30 am Q/A and Discussion Session with Anirudh Goyal
● 11:30 am - 11:35 am Wrap up

📌 SPEAKERS INFO
Anirudh Goyal (https://anirudh9119.github.io/) is a graduate student in CS at University of Montreal. He is a part of Mila (https://mila.quebec/en/), advised by Prof. Yoshua Bengio (https://yoshuabengio.org/). Before graduate school, he received a Bachelors in Computer Science at IIIT Hyderabad, where he worked on several research projects at CVIT under Prof. C.V Jawahar. He has also spent time at Google.

💙 –– THANK YOU ––
A big Thank goes to Dr. Rei Akaishi Sensei and his Lab Social Value Decision Making Unit, BTTC, RIKEN (https://cbs.riken.jp/en/faculty/btcc.svdm/) for patronizing Decision Intelligence Tokyo.