Sebastian Raschka on the state of tools for training, tuning, and evaluating machine learning models.
Sebastian Raschka is lead author of a new book from Packt entitled “Machine Learning with PyTorch and Scikit-Learn”. He is also an Assistant Professor of Statistics at the University of Wisconsin (Madison), and serves as the Lead AI Educator at Grid.ai.
We discussed a range of topics including:
- The state of educational resources and libraries for developers wanting to understand and use machine learning and deep learning.
- His recent research work in model evaluation tools, and modern methods for ordinal regression.
- Deep learning for structured data.
- Model tuning, model hubs, and data-centric AI.
Highlights in the video version:
- Introduction to Sebastian Raska
State of deep learning for developers
The need to identify use case and problem to solve before using deep learning
Domain expertise and models
Data Centric AI and tools
Drug Discovery Trials
Deep Learning for structured data
Graph neural networks, confidence intervals, and ordinal regression
Developing deep learning methods for statistical models
Spending time in a startup
Open source library for model evaluation
- A video version of this conversation is available on our YouTube channel.
- Jack Clark: The 2022 AI Index
- Zhe Zhang: How Technology Companies Are Using Ray
- Amit Sharma and Emre Kiciman: An open source and end-to-end library for causal inference
- Leo Meyerovich: The Graph Intelligence Stack
- Nic Hohn and Max Pumperla: Reinforcement Learning in Real-World Applications
- “Top Places to Work for Data Scientists”
- “Navigate the road to Responsible AI”
[Image: Cranium from Pixabay.]