Machine Learning: Key Battleground for Open Source Technology*
Despite all the attention and buzz, Machine learning(ML) is woefully overlooked in the community of free and open source technology. In this presentation, I will examine the still prevalent proprietary legacy of ML, introduce the current open source stack of ML development and applications, and evaluate new proprietary attempts entering ML. Then, I will share with you the strategy recipes that we may need, in a battle to keep the booming field of ML free and open source.
ML research has a long and still prevalent legacy of using proprietary technology. As ML becomes more and more popular with industry-scale large applications, many new and powerful proprietary pushes have been trying to enter different parts of the ML application stack. We need to be prepared for the challenges of keeping free and open source technology in ML, and the challenges come from both the legacy side and the new proprietary push side. However, ML is still overlooked in our conversations about free and open technology, although many open source alternatives have made ways into ML research, development, and industry applications.
1. Crash intro: ML stack in research, development, and applications
2. Before Open Source: ML’s proprietary legacy
3. Status quo in ML: where are we now?
4. Proprietary tech is coming(again): while we weren’t looking…
5. Strategies: what we can do, and your thoughts
Interest and knowledge in ML and enterprise-scale applications will be helpful, but are not required.
You will walk away with the big picture of many floating pieces in current ML development and applications. You will also learn about little-known historical context of ML development, its influence on the potential future of free and open source technology in ML, and some action recipes to keep ML free and open.
machine learning, open source technology, free software, proprietary legacy, challenge, research, Development, industry applications
Public speaking experiences in the past 5 years from smaller groups of 10+ people to 100+ people in different environments(e.g. educational, corporate, workshops, panels, etc.) in a range of topics from writing and diversity to coding, mathematics, and astronomy. This is a new talk.
Trained in Mathematics and Statistics, Helen is a data scientist and machine learning researcher with a passion for security, machine learning, and free and open source technology. Helen has worked on exciting projects in management consulting, technology start-ups, and non-profits in Asia, North America, and Europe. Helen has also taught mathematics, coding, writing, and astronomy to different audiences. When not writing code, building things, or sharing knowledge, she enjoys learning new spoken languages, photography, and long-distance running. She has many stories to tell, and you are welcome to ask her in person :)