Practical AI Escaping the "dark ages" of AI infrastructure
Dec 16, 2019
Evan Sparks, Co-founder and CEO of Determined AI, sheds light on the prevalent issues in AI infrastructure. He highlights why many are still in the 'dark ages' and discusses innovative solutions like fault-tolerant training and AutoML. Evan shares his journey from finance to AI, emphasizing the transition challenges. He also addresses the fragmentation in AI workflows and the need for cohesive systems that improve productivity and reproducibility. With an optimistic view, he explores the positive impacts of AI on the economy and environmental health.
AI Snips
Chapters
Transcript
Episode notes
From Finance to Infrastructure
- Evan Sparks's early machine learning work involved quantitative finance and NLP.
- He transitioned to infrastructure after realizing he spent more time on it than modeling.
Spark and Scalable Machine Learning
- Evan Sparks joined the AMP lab at UC Berkeley during the rise of Apache Spark.
- He contributed to MLlib and focused on building tools for large-scale machine learning applications.
Deep Learning's Infrastructure Gap
- Deep learning's rise brought new challenges, like long training runs and complex design decisions.
- Frameworks excel at model description and training but lack support for the broader workflow.