July 20–24, 2015
Portland, OR

Solve conference sessions

Harness the power of math to manipulate, secure, and create data.

1:30pm–5:00pm Tuesday, 07/21/2015
Carin Meier (Cognitect)
Clojure is an elegant, powerful language. With functional programming style and immutable data, we can solve the problem of concurrency that is very difficult to deal with in other languages. Read more.
10:40am–11:20am Wednesday, 07/22/2015
Kevin Burke (burke.services)
You might be good at designing coasters in "Roller Coaster Tycoon," but you could make even cooler coasters if you let Go build them for you. We'll look a little at RCT's code (written in x86) and how to reverse-engineer it. You'll learn how to design good genetic algorithms. Finally, we'll discuss the advantages of using Go's standard library for a project like this. Read more.
11:30am–12:10pm Wednesday, 07/22/2015
Joe Darcy (Oracle)
Learn about how floating-point arithmetic approximates real arithmetic, and lessons for more effective (and less surprising) numerical programming. Read more.
1:40pm–2:20pm Wednesday, 07/22/2015
Grant Ingersoll (Lucidworks)
Ever wonder how Watson beat all comers in Jeopardy or how Siri or Google Now work? Thinking about deploying question answering (QA) technology in your application? QA and NLP technology have finally hit the mainstream, and are making information access more powerful every day. The best part? Open source technologies make it easier than ever to build and deploy question answering technology! Read more.
2:30pm–3:10pm Wednesday, 07/22/2015
Ben Balter (GitHub)
Open source isn't open source without a license. GitHub is the de facto hub for creating and sharing open source software, but how much of it is truly open? How has license usage changed over time? How does licensing effect contribution, reuse, and project evolution? Join Ben Balter and Tal Niv, two of GitHub's legals for a quantitative analysis of license usage across all of GitHub's 19M repos. Read more.
4:10pm–4:50pm Wednesday, 07/22/2015
Tags: Java
Max De Marzi (NeoTechnology)
Do you need to build a recommendation engine like yesterday and have no idea where to start? How to get over the cold start problem? How to get some initial data? How do you know if its even working? Learn how to get past all that and get up and recommending quickly. Read more.
5:00pm–5:40pm Wednesday, 07/22/2015
Portland 252
James McCaffrey (Microsoft)
Swarm intelligence (SI) algorithms mimic the behaviors of groups such as flocks of birds and schools of fish. This session describes in detail four major SI algorithms: amoeba method optimization, particle swam optimization, simulated bee colony optimization, and firefly algorithm optimization. Attendees will receive Python source code for each algorithm. Read more.
10:40am–11:20am Thursday, 07/23/2015
Dawn Foster (Pivotal)
The real magic in any community comes from the people. I will show you tools and techniques for performing network analysis, to look at the people in your community along with the relationships between them. Why settle for boring numbers and line charts to describe your community when you can do cool visualizations that show how people connect within your open source community? Read more.
11:30am–12:10pm Thursday, 07/23/2015
Garen Torikian (GitHub)
The way math equations are written and represented have a long history that's woven into computer science. However, rendering math for the web has been a challenge. This talk explores the pursuit of rendering math beautifully for the web, culminating in the creation of a library that integrates with markup formats like Markdown and AsciiDoc. Read more.
1:40pm–2:20pm Thursday, 07/23/2015
Tags: Python
A boyle (New Relic)
Learn how digital signal processing can help you look at data, and even the world around you, with a new critical eye. Gain an understanding of the methods behind speech recognition, image manipulation, data compression, and more. Read more.
2:30pm–3:10pm Thursday, 07/23/2015
Carin Meier (Cognitect)
Take a step back from your normal programming approach and discover a new way of looking at problems. All living organisms' information systems are based on chemical processes. What can we learn by using this metaphor of chemistry in our programming? Read more.
4:10pm–4:50pm Thursday, 07/23/2015
Patrick Barton (O'Reilly School of Technology)
This talk addresses improving the ability of hierarchal temporal memory (HTM) to predict electricity demand, helping the algorithm by providing some of the complementarity data streams currently applied to demand analysis, and including some goodness-of-fit metrics that address known characteristics of electric load. Read more.
5:00pm–5:40pm Thursday, 07/23/2015
Matthew Taylor (Numenta)
What if you had software that tracked location history so well it could remember all the walks you take, all your commutes, and even the quests you take in your favorite video games? With this level of detail captured and a bit of algorithmic brilliance, NuPIC allows for instant anomaly detection, and in the process, opens up a whole new world of intelligent applications for pattern recognition. Read more.
10:00am–10:40am Friday, 07/24/2015
Mocha.jl is an efficient and flexible deep learning framework for Julia. It supports multiple computation backends, leading to 20~30 times faster training on a modern GPU device. We will use an example to illustrate the user interfaces of Mocha.jl and also introduce the design and architecture behind the library implementations. Read more.
11:10am–11:50am Friday, 07/24/2015
Alasdair Allan (Babilim Light Industries), Paul Fenwick (Perl Training Australia), Jon Manning (Secret Lab), Paris Buttfield-Addison (Secret Lab), Tim Nugent (lonely.coffee)
Join the authors of "The Kerbal Book" on a panel where they regale you with tales of their adventures in the Kerbal Space Program, the increasingly popular and disturbingly realistic space programme simulator game enjoyed by geeks around the world. Learn how and why you should go to space, and what you can learn from it! Science will be involved. Read more.