Skip to content

Instantly share code, notes, and snippets.

View Sachin-A's full-sized avatar

Sachin Ashok Sachin-A

35 followers · 41 following
View GitHub Profile
@Sachin-A
Sachin-A / deltaInductionsTask2.md
Created July 9, 2018 20:32
Task #2 in ML for delta inductions (Open Profile)

ML Task #2

Problem Statement:

Using clustering for some unsupervised learning!

Clustering is the task of grouping a set of objects in such a way that objects in the same group (called a cluster) are more similar (in some sense) to each other than to those in other groups (clusters)

For this task, there will not be a specified dataset. The selection of dataset is up to you but a good selection of dataset to showcase the power of clustering will be appreciated.

@Sachin-A
Sachin-A / deltaInductionsTask1.md
Last active June 17, 2018 15:39
Task #1 in ML for delta inductions (Open Profile)

ML Task #1

Problem Statement:

Using linear regression for rank predictions!

In this type of machine learning task, the objective of the computer program is to predict a numerical value given some input. Linear regression models can often be fitted using the least squares approach but they can also be fitted in other ways!

Keeping that in mind and in the spirit of the FIFA football world cup finally kicking off, it's time to make some rank predictions!

@Sachin-A
Sachin-A / resources.md
Created June 17, 2018 15:26
Machine learning resources for beginners
@Sachin-A
Sachin-A / breaking_recaptcha.md
Last active October 10, 2017 17:36
Summary of ReCaptcha paper
@Sachin-A
Sachin-A / argon2.md
Last active October 26, 2016 09:11
Overview of Argon2: A memory hard function for password hashing

Argon2

  • Argon2 is a key derivation function that was selected as the winner of the Password Hashing Competition in July 2015. It was designed by Alex Biryukov, Daniel Dinu, and Dmitry Khovratovich from University of Luxembourg.
  • It uses the BLAKE2 hash algorithm to securely scramble input data (password and salt).

Introduction

Problems with existing designs

  • Should the memory addressing (indexing functions) be input-independent or input-dependent, or hybrid?
  • Is it better to fill more memory but suffer from time-space tradeoffs, or make more passes over the memory to be more robust?
  • How should the input-independent addresses be computed? Several seemingly secure options have been attacked.