# Using Elixir for Data Science and Machine Learning

**URL:** <https://forum.elixirforum.com/t/using-elixir-for-data-science-and-machine-learning/849>\
**Category:** Discussions\
**Tags:** machine-learning, data-science\
**Created:** [June 13, 2016, 8:47am UTC](https://forum.elixirforum.com/t/using-elixir-for-data-science-and-machine-learning/849 "2016-06-13T08:47:11Z")\
**Posts on this page:** 1\
**Showing post:** 31

<div class="post-metadata">

**Author:** ![ericsteen](https://forum.elixirforum.com/letter_avatar_proxy/v4/letter/e/3ab097/32.png) [@ericsteen](https://forum.elixirforum.com/u/ericsteen)\
**Post date:** [July 25, 2020, 6:05pm UTC](https://forum.elixirforum.com/t/using-elixir-for-data-science-and-machine-learning/849/31 "2020-07-25T18:05:40Z")

</div>

Let’s keep in mind that supervised learning is only one approach to ML, and backpropagation (the heaviest number crunching part of Deep Learning) is [falling out of favor](https://www.i-programmer.info/news/105-artificial-intelligence/11135--geoffrey-hinton-says-ai-needs-to-start-over.html). DL is not very scalable, and Unsupervised Learning (i.e. Reinforcement Learning) seems to be expected to be the key path to AGI (artificial general intelligence) — according to [Richard Sutton](https://en.wikipedia.org/wiki/Richard_S._Sutton) and others

With that in mind, I have been successfully using [Matrex](https://github.com/versilov/matrex) for high-dimensional vectorized computation for Multi-Armed Bandits (Elementary Reinforcement Learning). You can check it out [here](https://github.com/upstarter/automata/blob/master/lib/automata/automaton_types/reinforcement_learning/bandits/mab.ex). Its part of [The Automata Project](https://www.github.com/upstarter/automata). Down the road I may need to use python or julia via erlports (or maybe even docker containers as used [here](https://youtu.be/p58sFfgMvdI?t=1581)) for the vectorized parts, but for prototyping, things are going well so far.

In my view, [Neuroevolutionary Typology and Weight Evolving Artificial Neural Networks (TWEANN)](https://www.youtube.com/watch?v=TcUqGIHq8rA&list=PLmgEqxC4Fyq1fTBdJeqGOij3jgb-5gvQx&index=89&t=0s) with [Novelty Search](http://eplex.cs.ucf.edu/noveltysearch/userspage/) is one of the most promising alternative approaches, and Elixir has a head start in this sub-field of ML thanks to Gene Sher’s book.

As far as python interop goes, something like [this](http://davekuhlman.org/elixir-poolboy-erlport-datasci.html) looks pretty appealing for scaling ML as well.

[The Automata Project](https://www.github.com/upstarter/automata) is seeking contributors if anyone here is interested.

---

_[View the full topic](https://forum.elixirforum.com/t/using-elixir-for-data-science-and-machine-learning/849)._