SCAN in Tensorflow

A Tensorflow implementation of DeepMind's Symbol-Concept Association Network (SCAN: Learning Abstract Hierarchical Compositional Visual Concepts).

SCAN is a framework for learning concepts paired with visual primitived. By teaching logical operators to SCAN, it even can learn to imagine new concepts from logical operators. See the paper and DeepMind blog for more details.

Usage

Prerequisites

• Python 3.4
• DeepMind Lab
• Tensorflow 1.2

This is my configuration, other versions might work, too. If don't, let me know.

It is worth mentioning that the origin DeepMind Lab repo hasn't been updated for months, and it doesn't support python3. On the other hand, I'm used to work on my Gentoo Linux with Python3. In order to make it work, I have forked this repo to my github and make it Gentoo and python3 supportable. If you are also using an unsupported platform or environment, see here for details.

Instructions for runing

$ git clone https://github.com/naturomics/SCAN-tensorflow.git
$ cd SCAN-tensorflow

$ python main.py

Results

Author

Naturomics Liao - @naturomics