Learning Resources Recommended by the UCL DSS Team

Committee


Sitting at home in quarantine? Want to learn something new?

We hope all of you are well, and that the time leading up to the exam term hasn't been

too stressful. We extend our well wishes for everyone affected by the pandemic, and a

thank you to everyone who has contributed in one way or another.While there might be

disruptions to Summer plans, now is a prime opportunity for learning and building on

your skills! At DSS, we've compiled a list of websites and courses that we think would be

beneficial for all of you.

NOTE: We are working actively with our science team and external education platforms

to secure an exclusive deals for DSS members, so please look out for further updates on

that.


Data Science Resources

The following courses and resources can provide you with a solid knowledge

foundation in machine learning with theories and praticals.

Coursera - Machine Learning by Stanford

This is a great introductory course by the top lecturer at Stanford University and Coursera

Andrew Ng. It provides a rigorous approach towards the mathematical intuition behind

machine learning algorithms, with hand-on implementation.

Coursera - Data Science Specialisation by Johns Hopkins University

HarvardX - Data Science

Introduction to Deep Learning by MIT

GitHub - inst0060-notes by Tony Wu

This is the course notes for INST0060: Foundation of Machine Learning and Data

Science module taught at the Department of Information Studies at the UCL, with

rigorous proof sketches to many math intuitions in Supervised Learning, Reinforment

Learning, Unsupervised Learning and Neuro Networks algorithms.


Mathematics, Statistics and Computer Science Resources

Data Science is a product of mathematics, statistics, computer science and

programming. These resources provide you with the essential toolsets for discovering

the mechanisms behind each algorithms from a mathematician, statistician and

computer scientist's perspective

MIT - MIT Open Course Ware (OCW)

A collection of the open courses provided at one of the best STEM-related universities:

MIT. The 18.0x series of introductory courses by the Mathematics Department and the

6.0x series of introductory courses by the Computer Science Department are highly

recommended. The courses cover a wider and deeper range of contents than some

similar courses taught at the UCL with courseworks, examinations, projects and sample

solutions.

Some recommended courses:

18.02 Multivariable Calculus

18.03 Differential Equations

18.05 Introduction to Probability and Statistics

18.06 Linear Algebra

18.062J Mathematics for Computer Science

18.065 Matrix Methods in Data Analysis, Signal Processing, and Machine

Learning

6.041SC Probabilistic Systems Analysis and Applied Probability

6.005 Software Construction

6.087 Practical Programming in C

6.088 Introduction to C Memory Management and C++ Object-Oriented

Programming

Note that:

Professor Strang's lectures on 18.06 and 18.065 are frequently regarded as the

"best Linear Algebra course". His textbooks are used as the standard textbooks for

many univerisities around the world. He published also a new book in 2019 on the

linear algebra for machine learning. Machine Learning is highly related linear

algebra so these are definitely worth checking out

Coursera - Algorithms by Princeton University

This is a 2-part course on algorithms by Sedgwick and Wayne, professors at Princeton.

The supplementary reading Algorithms, 4th Ed is regarded as the Algorithms Bible by

many. The course provides you with a thorough introduction to many algorithms and the

mathematical proof sketches of the running time and memory usage efficiencies behind

each algorithm. The Java implementation of data structure and Sedgewick's interfacr-

orientated approach for Java Implementation is absolutely inspiring. Upon the

completion of the course, you will not only be able to implement a range of data

structures and algorithms on your own but also become a master in Java and the

mechanism behind it.


Coding Resources

Scrimba

It's an interactive way to learning to coding. It provides live screencast so that you can

pause at anytime to play around with the freshly-written codes by the lecturer in real-

time in the browser with live rendering, compiling and interpreting enabled, which frees

you from setting up your local environment. Courses on front-end engineering, data

structure, and programming are avaible.

LeetCode and HackerRank, your humble old friend for cracking the code interview.


Books

Cracking the Coding Interview

The Hundred-Page Machine Learning Book

Algorithms, 4th Ed

Introduction to Linear Algebra, 5th Ed

Linear Algebra and Learning from Data

Back to Index →