07-14, 14:35–15:05 (Europe/Dublin), Liffey Hall 1
The use of transfer learning has begun a golden era in applications of Machine Learning but the development of these models “democratically” is still in the dark ages compared to best practices in Software Engineering. I describe how methods of open-source software development can allow models to be built by a distributed community of researchers.
Here, I elaborate on why we should develop tools that will allow us to build pre-trained models in the same way that we build open-source software. Specifically, models should be developed by a large community of stakeholders who continually update and improve them. Realizing this goal will require porting many ideas from open-source software development to building and training models, which motivates many threads of interesting research and opens up machine learning research for much larger participation.
none
Expected audience expertise: Python –some
Abstract as a tweet –The use of transfer learning has begun a golden era in applications of Machine Learning but the development of these models “democratically” is still in the dark ages compared to best practices in Software Engineering. I describe how methods of open-source sof
Steven Kolawole has his technical skillset cuts across Data Science and Software Engineering, with a bias for ML Research these days. His research interests focus on resource-efficient machine learning in terms of computational resources and low-resource/limited labeled data.
He is and has been heavily involved in varieties of ML subfields including ML Engineering, Software Engineering, Data Engineering, Data Science/Analytics, and Cloud Computing.
Steven is also big on knowledge sharing via community mentorship and collective growth, open-source development, meetups facilitation, speakership, technical writing, research, and he gets kicks from helping tech muggles find their feet.