ELL 881 2018: Fundamentals of Deep Learning
Instructors: Raghavendra Singh @raghavsi & Vineet Kumar @vineetm
Teaching Assistants (TAs): Aravind @Maxaravind & Vinay @VinayKyatham
Master the fundamentals of Deep Learning and implement the ideas learnt in Tensorflow.
We will closely follow the Deep Learning Book
Relevant concepts would be implemented using Tensorflow
Relevant low-level and important details such as data handling and manipulation using tf.data would be taught
Individual Project: Implement two selected publications pertaining to NLP and Computer Vision. Further details TBD later in the course
Lecture | Contents | Readings |
---|---|---|
Lec 01 (27 July) | Introduction; Logistics; Linear Algebra; Probability & Information Theory | Ch 01, 02, 03 |
Lec 02 (03 Aug) | Numerical Computation; Machine Learning Basics | Ch 04, 05 |
Lec 03a (08 Aug) | Deep Feedforward Networks | Ch 06 |
Lec 03b (10 Aug) | Computational Graphs; Backprop | Ch 06 |
Lec 04a (29 Aug) | Tensorflow, Regularization | |
Lec 04b (31 Aug) | Regularization | |
Lec 05a (05 Sep) | NLP Basics, NLP data pipeline | Stanford CS230 Blog |
Lec 05b (07 Sep) | Word2Vec | Paper,Blog |
Lec 06a (12 Sep) | Word2Vec (Contd.) RNN Basics, Language Modeling | Ch 10 |
Lec 06b (14 Sep) | RNN (Contd.), Perplexity, Gradient Clipping, Checkpointing, Project 1 | Ch 10 |
Project 1 START | 14 Sep 2018, 05 pm | Project1 |
Project 1 END | 12 Oct 2018, 01 PM | |
02 & 09 Nov | Project 1 In-Class Viva | |
11 Nov | Mandatory submission of current version of Project 2 | |
14 Nov | Lecture on GANs | |
18 Nov | Project 2 End. Try to submit by 16th night (before majors) | |
26-28 Nov | Project2 Viva on one of these days (TBA) |
Content | Weightage |
---|---|
Project | 45 % |
End Term Exam | 20 % |
Assignments | 15 % |
In-class Quiz; Class Participation | 10 % |
Minor-I,II | 10 % |