arrow_back Machine Learning on Kubernetes using Kubeflow
Serverless Machine Learning Platform arrow_forward
Introductory workshop on Computational Machine Learning
Submitted by Abhijith Chandraprabhu (@abhijithc) on Tuesday, 1 May 2018
Technical level: Intermediate
You have been hearing about machine learning (ML) and artificial intelligence (AI) everywhere. You have heard about computers recognizing images, generating speech, natural language, and beating humans at Chess and Go. In this workshop, you will be learning the math and program the math of AI, for example we start by hand coding all the components of a neural network, without calling any libraries. The concepts will be taught in Julia, a modern language for numerical computing and machine learning. We will eventually use Flux, which is a Julia machine learning stack, written 100% in Julia, to build some deep learning applications like image recognition and language detection.
The takeway from this workshop is good understanding of the first principles of deep learning and ability to build elegant high-performance deep learning applications.
- Some Motivations for Machine Learning
- Data - Key to AI and ML
a. Computer Representation of Data
- Modeling Data
a. Functions and parameters b. Data fitting by varying parameters c. Quantifying how far we are from the goal: the loss function
- What is Learning
a. Motivation: Fitting Parameters By Hand b. “Learning by Nudging”: The Process of Descent¶
- Introduction to Neurons
- Build neural networks
- Not to shy away from getting into some mathematical concepts
- Commitment to strive towards understanding the concepts and program for applications
- Regular attendance and timely completion of assignments
- Active participation in the classes
- Commitment to follow on work or projects in order to apply the concepts in real life
Abhijith is currently working as a data scientist at Julia Computing, Bangalore. Previous to that he worked as data scientist at Gramener, Hyderabad.