Q n A with Dr Tom Starke
summary
Claims that fully autonomous trading is ready are false; research shows progress on narrow signals but not end-to-end AI trading because markets contain heavy noise and changing rules that models break on deployment. Demonstrated use cases include satellite car counts, lid-level oil storage imagery and sentiment signals. Practical tooling centers on Quantopian platform for backtesting with provided data, Python (and C++ for production), and data sources like Quandl and FRED; commercial market feeds often cost >$100k/year.
AI-written summary · about
related talks
30mHow to build a stock market money printing machine with AI
16mOperationalizing AI in Finance: Real-Time Analytics
19mNow in AI - How We Got Here
59mAI in Finance: Scope and Examples- 59mAI in Finance: Scope and Examples
44mArtificial Intelligence Reloaded - AI Apps in the Industry
18mKeynote: AI for Financial Services
47mQuantitative Trading with Machine Learning