Past Meetup

Edmonton Data Science Meetup

This Meetup is past

105 people went

Location image of event venue

Details

We are proud to announce the 5th Edmonton Data Science meetup on October 5 that is jointly sponsored with IEEE Com/Comp Society (http://northerncanada.ieee.ca/).

The presentations include:

• "Data Science at Mitre Media" - Filipe Mesquita (Vice President of Data Science @ Mitre Media)

• "A talker on Docker: How containers can make your work more reproducible, accessible, and ready for production" - Finbarr Timbers (Analyst @ Darkhorse Analytics)

• "Evaluating Real-Time Strategy Game States Using Convolutional Neural Networks" - Marius Stanescu (PhD Student @ CS UofA)

Pizza will be provided before presentations and after presentations we will head to a pub for free beer and networking!

More info about the talks:

Data Science at Mitre Media

Filipe will discuss the challenges of creating a Data Science department from scratch. In particular, he will discuss: (1) what Data Science looks like in a media company, (2) how to propose and prioritize Data Science projects and (3) when to give up on a project.

A talker on Docker: How containers can make your work more reproducible, accessible, and ready for production.

Containers are a popular technology that wrap up code with everything it needs to run: runtime, system tools, system libraries – anything you'd normally install on a server. As a result, your container runs the same everywhere- no more "Well, it works on *my* machine." Containers are particularly useful in situations that require a large number of complex or fragile dependencies, or in scenarios where the developer has no idea of what the production environment looks like. Data science often has both problems. The talk will go over a number of problems that the presenter faces regularly at work, and will discuss how containers help solve them.

Evaluating Real-Time Strategy Game States Using Convolutional Neural Networks

Real-time strategy (RTS) games, such as Blizzard’s StarCraft, are fast paced war simulation games in which players have to manage economies, control many dozens of units, and deal with uncertainty about opposing unit locations in real-time. Even in perfect information settings, constructing strong AI systems has been difficult due to enormous state and action spaces and the lack of good state evaluation functions and high-level action abstractions. To this day, good human players are still handily defeating the best RTS game AI systems, but this may change in the near future given the recent success of deep convolutional neural networks (CNNs) in computer Go, which demonstrated how networks can be used for evaluating complex game states accurately and to focus look-ahead search. In this paper we present a CNN for RTS game state evaluation that goes beyond commonly used material based evaluations by also taking spatial relations between units into account. We evaluate the CNN’s performance by comparing it with various other evaluation functions by means of tournaments played by several state-of-the-art search algorithms. We find that, despite its much slower evaluation speed, the CNN based search performs significantly better compared to simpler but faster evaluations. These promising initial results together with recent advances in hierarchical search suggest that dominating human players in RTS games may not be far off.

----------

Edmonton Data Science (EDS) meetups are a platform for data science enthusiasts and professionals to:

• learn about big data techniques and tools

• discuss best practices in big data and data science

• promote big data and data science among university students and the local tech community

• learn about practical applications of data science methods

• learn about companies that use data science as the foundation of their business processes

We are passionate about data science and big data. We hope to create a vibrant data science community that is engaged in sharing and learning different topics in the field.

In these meetups, we intend to a) share best practices and knowledge of data science-based startups and corporations with university students, b) introduce research works on big data and data science to the local tech community. We aim to connect the students to companies and make them more aware of our vibrant local tech community. We also invite professors and researchers to present their work on data mining and machine learning methods to the professional community. Each meetup will include a presentation introducing a company who is using data science techniques, and two presentations about big data, data science and machine learning techniques.