Back to Results

EFTA02478888.pdf

Source: DOJ_DS11  •  email/external  •  Size: 75.8 KB  •  OCR Confidence: 85.0%
View Original PDF

Extracted Text (OCR)

From: George G. Mueller < Sent: Monday, November 23, 2015 6:17 PM To: jeffrey E. Cc: Gino Yu Subject: Meeting Followup Hey Jeffrey, it was a pleasure meeting you on Friday. I found the conversat=on enjoyable though far too fleeting for my liking. I hope to carry on our=dialog about companies and markets. Are you interested in hearing more abo=t Cerebellum Capital (see below) our machine learning fund? It think there is some nice crossover between y=ur knowledge and our mission. Let me know if you'd like to hear more abo=t this. Cheers! -Geo Cerebellum Capital is a hedge fund m=nagement firm whose investment programs are continuously designed, execute=, and improved by a software system based on techniques from statistical m=chine learning. The system is responsible for consta=tly creating its own new models for how the markets will move, testing tho=e models, refining them, and learning trading strategies that take advanta=e of these predictive models. The system is provided with a wide variety of traditional and non-traditional,=publicly available and licensed data streams as inputs to its model creati=n and improvement process. Cerebellum's software system learning optimizes=for a proprietary mix of expected return maximization, risk/volatility reduction across the portfolio, and p=rtfolio independence from major markets when they trend downward. Cerebell=m's architecture for continuous improvement, self-diagnosis, and fault tol=rance is based on a collective 30 years research in the area of statistical machine learning applied to r=al world, mission critical time-series problems. 1 EFTA_R1_01593931 EFTA02478888

Document Preview

Extracted Information

Dates

Document Details

Filename EFTA02478888.pdf
File Size 75.8 KB
OCR Confidence 85.0%
Has Readable Text Yes
Text Length 1,666 characters
Indexed 2026-02-12T18:03:34.324020

Related Documents

Documents connected by shared names, same document type, or nearby in the archive.

Ask the Files