The TWIML AI Podcast (formerly This Week in Machine Learning & Artificial Intelligence) Panel: The Great ML Language (Un)Debate! - #393
Jul 20, 2020
Guest
Chris Lattner
Guest
Avi Bryant
Guest
Gabriella DeCuroz
Guest
Catherine Nelson
Guest
Robert Osizu-Aness
Guest
Huda Nassar
Guest
Barack Canberr
Guest
Chris Nurenberger
In a lively debate, Chris Nurenberger, a machine learning expert, champions Clojure for its conciseness. Barack Canberr pushes for JavaScript's accessibility, while Huda Nassar highlights Julia's speed and community. Robert Osizu-Aness discusses probabilistic programming's potential in NLP. Catherine Nelson emphasizes Python's flexibility, and Gabriella DeCuroz celebrates R's supportive resources. Avi Bryant discusses Scala's challenges, and Chris Lattner touts Swift's performance. Together, they explore the strengths and weaknesses of various programming languages in the ML landscape.
AI Snips
Chapters
Transcript
Episode notes
Democratizing Machine Learning
- Machine learning should be accessible to programmers of all backgrounds and language expertise.
- Focusing on algorithm implementation is key, as language is just a tool.
JavaScript Journey
- Barack Canberr started in machine learning with PHP, facing criticism for the unconventional choice.
- He then chose JavaScript, initially ironically, but found he could combine algorithms, visualization, and interactivity seamlessly.
Multiple Dispatch in Julia
- Multiple dispatch in Julia allows specializing functions for new types easily.
- This streamlines code reuse and collaboration by centralizing code on platforms like GitHub.