Stochastic Learnings

Random walk in the land of machine learning, algorithms, and software engineering.

Follow me on GitHub

20 Dec 2018

Our paper has been accepted to ICLR 2019!

Woo hoo, New Orleans, here I come! Just got an email from the organizers that our paper has been accepted to the ICLR conference. Of course, all credit goes to Saeed – I just helped with the engineering part and ran a few experiments. I really hope that we’ll be able to open source the code soon – it has some modules that can be useful for domains other than SAT. Meanwhile, here’s the paper:

Learning To Solve Circuit-SAT: An Unsupervised Differentiable Approach

18 Dec 2018

My impressions from NeurIPS 2018 conference

My team has a great tradition: everyone who goes to a major conference must write a report about it and share his/her impressions with the team. In that report we usually list the most interesting papers to be studied later in our ML reading group meetings. So the list below is effectively my study plan for the next 6 months.

Read more: My impressions from NeurIPS 2018

29 Oct 2018

Analyzing Linux object files

Object files in Linux come in four different flavors:

  • .o object file. This is the most basic object file format in Unix. There is usually a 1:1 correspondence between a source file, e.g. in C, C++, assembly, or Fortran.
  • .a static library. A static library in Linux is just a collection of .o object files archived into one file. That is, .o file to .a file is like .class file to .jar file in Java.
  • .so dynamic library.
  • An executable (usually in ELF format).

In this tutorial you will learn how to introspect all four types of binary files.

Read more: Introspecting Binary Files in Linux

23 Oct 2018

pytorch-annotated: MNIST Example Part 2

In the second part of my PyTorch MNIST example annotation I cover the forward propagation and torch.nn components used in the example.

Here’s the link to the python notebook: Model components and forward propagation.

22 Oct 2018

pytorch-annotated: MNIST Example Part 1

I’ve finished the first part of PyTorch MNIST example annotation. It covers the data loader and related modules.

Here’s the link to the python notebook: Loading the data.

19 Oct 2018

pytorch-annotated project

I’ve started a small project to document PyTorch internals. Eventually I hope to cover low-level APIs, like ATen and good old TH.

I will start with a simple python notebook that explores the components of the PyTorch MNIST example.

Here’s the new git repository: pytorch-annotated.