You Don't Know What "dropout" Is: Churn-Creating Cutting-Edge Tech
summary
A Reface team member examines why many practitioners use techniques like dropout without understanding them and argues that breakthrough ML requires deep grasp of fundamentals. The talk describes Reface’s ML practices, barriers that prevent engineers from truly understanding their methods, and how to systematize production of new approaches at scale.
AI-written summary · about
chapters (7)
- 0:00Provocative title
- 6:50Fashion tech innovation
- 11:47Dropout definition
- 16:46Dropout math and impact
- 20:58Dropout use cases
- 26:02Innovating without papers
- 27:28Q&A
related talks
05m5 concepts I took from Data Science to Team Leading
28mUnravelling Hidden Technical Debt in ML: A Pythonic Approach to Robust Systems
48mPrinciples and Practices of Building Machine Learning Systems
27mFireside Chat
46mI know Machine Learning but... am I ready for the real world?
06mInterview
39mChanging Realities - The Final Frontier for Machine Learning?
45mBuilding the future, a founder's perspective