Abstract: Academics and practitioners alike believe that continual learning (CL) is a fundamental step towards artificial intelligence. Continual learning is the ability of a model to learn continually from a stream of data. In practice, this means supporting the ability of a model to autonomously learn and adapt in production as new data comes in. The idea of CL is to mimic humans ability to continually acquire, fine-tune, and transfer knowledge and skills throughout their lifespan. CL of models in production will improve accuracy, and bring artificial intelligence one step closer to real human intelligence.
In this presentation, Yochay Ettun will give a detailed explanation of why CL is important, how it improves model accuracy, and how to implement CL in practice using TensorFlow, Kubernetes and cnvrg.io. Key takeaways include:
- Understanding of Continual Learning
- Optimizing for accuracy with CL
- How to use TensorFlow, Kubernetes, and other tools to apply CL
- How to make automatically adaptive machine learning
- Adapting to shifting data distributions
- Coping with outliers
- Retraining in production
- Adapting to new tasks
- Deploying ML pipeline to production
Bio: Coming Soon!