About the reading group

Overview

A Reading Group will meet on alternate weeks. It aims to cover advanced topics in deep learning ranging from empirical risk minimization to reinforcement learning. Participants may be expected to master and present the main ideas from an important paper. The sessions are meant to be interactive.

  • The group reads two or three recent research papers per session.
  • Each paper is prepared and presented by two participants, one defends the paper, the other tries to criticise it.

Questions related to the papers and the reading group are being discussed on slack. You should be able to access it with your @cam.ac.uk account.

Main objectives

  1. Read and review deep learning papers.
  2. Learn to present, defend and criticise a research paper

Prerequisites

This is an advanced reading group for researchers and students at late stages of their studies. The knowledge of the material of the following courses is therefore expected:

  • Applied Probability
  • Statistics IB
  • Principles of Statistics
  • Markov chains
  • Optimisation

It is also preferred that you are familiar with at least one deep learning package

31 October

The papers to read are listed at talks.cam.ac.uk

14 November

The papers to read are listed at talks.cam.ac.uk

Some tips:

  • Start reading your assigned papers early and prepare a few slides/discussion
  • Think about how what you read fits with other findings
  • Question any decision and claim made by the authors

    Have a story and motivate well. Feel free to make the presentation interactive 😉

    Expected workload is 4-6 hours per paper