Monday, January 15, 2018

Nice Lecture series on Causality

In Reinforcement Learning you need to define functions for rewards and state transitions.   Big data is an good source of modeling external actors to a deep learning system.  What is needed is a causality model for the data to be able to simulate the external world correctly.

This set of lectures are quick introduction to the field.

part 1

part 2

part 3

part 4

Friday, April 21, 2017

On back propagation algorithm



Nice detailed explanation of back propagation in neural networks:






Sunday, June 12, 2016

Calculus and Algebra

I thin this video capture the essence of calculus.



"We have a geometric concept, and we want an algebraic expression for it".

It could be: we have XYZ concept, and  want and algebraic expression for it.

An XYZ calculus (such as Tensor calculus in case of Geometry, differential calculus in case of algebra, Relational Calculus in case of Relational Algebra, Lambda Calculus in case of Function algebra...)  is  used to write an algebraic expression which is then solved with algebraic laws to reason.  

The method seems to always be "Write down the correct identity"...  Struggling to find how this applies to Lambda Calculus..

Sunday, October 4, 2015

Tuesday, December 30, 2014