PyData Montreal May Meetup

Join us for our 4th PyData Montreal meetup.
This time we will have the following speakers:
- Simon Ouellette from Pointus Partners
- Peadar Coyle, one of the PyMC3 contributors
- Fred Mailhot from Autodesk

----------------------------------------------------------------------------------------------------
**Agenda**
6:00 pm—Doors open

6:10 pm— "Introduction to probabilistic programming with PyMC3" by Simon Ouellette

We will introduce the topic of probabilistic programming and show concrete examples of stochastic modelling from the grounds up, with an emphasis on time series.

About Simon:
Simon Ouellette is a quantitative trader, currently spearheading the AI trading initiative at Pointus Partners. Starting his career as a computer scientist about 15 years ago, he increasingly gravitated towards data science-related roles, culminating in his current position. He also creates content about topics such as stochastic modelling, reinforcement learning and quantitative trading via his blog at http://www.simonouellette.com/

6:50 pm—Break

7:00 pm—"Lessons learned from PyMC3" by Peadar Coyle

Building an Open source project is hard. Peadar will share his own experiences and lesson learned from evangelising, building code and implementing new algorithms and cutting edge research. Listeners will leave with a better understanding of how a successful OSS project works.

About Peadar:
Peadar is an entrepreneur and OSS contributor based in London. He's now working on an AI startup that aims to produce a personal audio assistant for educational purposes. He's worked with companies such as Amazon and Zopa and has spoken at PyData/PyCon conferences around the world.

7:15 pm—Break

7:25 pm—"Learning Vowel Harmony with RNNs" by Fred Mailhot

Most languages across the world demonstrate restrictions on which sounds can co-occur, either in immediate adjacency, or within a word. In this talk, I’ll describe a particular set of co-occurrence restrictions called vowel harmony, which is of interest to linguists, and discuss some personal work modeling their acquisition with so-called sequence-to-sequence networks.

About Fred:
Fred Mailhot is Lead Data Scientist at Autodesk Canada, in the Data Analytics and Automation division. His academic background is in Linguistics and Computer Science, and he's been doing Data Science and NLP in industry for 7 years. He wishes there was more time to read.

8:10 pm—Break and networking