Generalization and Specialization

Graybox Trading

Dec 28

My intuition for markets is very good. I have a high generalized ability to read markets. It is less likely that I know the specific specialized optimal parameters for my trades. Specialization is where machine intelligence such as programming and optimization already excel. And, machine learning is starting to offer the promise of generalized intelligence.

However, this generalized intelligence while being more human-like is not exactly the same. It is not a surprise that computers are more intelligent then humans in specific ways. Calculators exceeded human intelligence in the specific objective of counting and mathematical operations, likely since invention. I believe the AI revolution is real: however, I do not think it is on the path to higher intelligence, yet.

The fact that machine intelligence is different, in context of trading, might be a virtue instead of a just a risk. For example, deep learning networks have been fooled into giving high confidence scores for image classification on images that we identify as only noise. Some suspect that algorithmic intelligence was superior to humans when it came to trading. However, it might simply have been that it was a new intelligence, a different intelligence. Not a better or worse intelligence but superior by virtue of being new and unique.

My goal in my trading for 2019 is to take the best aspects of my existing strengths while replacing my weaknesses with algorithmic intelligence.

About the Author

The author is passionate about markets. He has developed top ranked futures strategies. His core focus is (1) applying machine learning and developing systematic strategies, and (2) solving the toughest problems of discretionary trading by applying quantitative tools, machine learning, and performance discipline. You can contact the author at