502 reputation
59
bio website
location Chicago, IL
age 30
visits member for 3 years, 6 months
seen Dec 20 '13 at 17:37

My name is Kiril Gantchev, I'm a software engineer and my interests include: machine learning, artificial intelligence, bitcoin, trading and making the world a better place!

I blog about technology in general (mostly things I encounter):
codesprout.blogspot.com

Sometimes I blog about ML/AI stuff:
mlai-lirik.blogspot.com

P.S. If you're wondering, my gravitar image is an endogenous retrovirus.

kgan...@gmail.com


Jan
31
awarded  Yearling
Aug
10
awarded  Nice Question
Mar
12
awarded  Nice Answer
Feb
7
awarded  Beta
Feb
7
accepted How to combine various equity measures into a single measure (vector magnitude)
Feb
6
comment How to combine various equity measures into a single measure (vector magnitude)
@shabbychef, I'm trying to figure out a way to "collapse" the vector so I don't have to compare each metric separately. I have thousands of instances to compare and I want to do it as fast as possible. Comparing 5 dimensions is going to be much slower than just comparing a single value.
Feb
6
asked How to combine various equity measures into a single measure (vector magnitude)
Feb
4
awarded  Scholar
Feb
4
accepted How are risk management practices applied to ML/AI-based automated trading systems
Feb
4
awarded  Commentator
Feb
4
comment How can I go about applying machine learning algorithms to stock markets?
@Neil: semantics... maybe it's not the general consensus, but the attitude towards AI is that it's voodoo science and you try it if you have $800 million to blow. I don't have such an attitude, but when I talk to various people I get that vibe.
Feb
3
comment How are risk management practices applied to ML/AI-based automated trading systems
@shabbychef: If it was easy, everybody would be doing it... some people do give up. The assumption is that your trader (or algorithm in this case) is pretty good, but you need proper risk management to establish correct lot sizes, track exposure and controlling losses. The risk management you're talking about does not address those things.
Feb
2
comment How are risk management practices applied to ML/AI-based automated trading systems
@shabbychef, OK, you're thinking of a completely different architecture for an ML system...
Feb
2
comment How are risk management practices applied to ML/AI-based automated trading systems
@shabbychef, like I said: there is no backtesting! If you hire a trader you don't backtest the trader to see how they're going to do, instead you let them trade with a paper trading account and once you see that they're doing good you actually let them trade with real money. You take the same approach with ML: you train your ML with a rolling training set, once you reach sufficient performance you start paper trading and that is effectively your "backtest." The training does not stop even if you start trading with real money, that's why you need a separate measure for risk.
Feb
2
comment How are risk management practices applied to ML/AI-based automated trading systems
@shabbychef, online machine learning eliminates the use of back-testing. The point of online machine learning is that you never stop learning: your machine learner constantly generates viable candidates (i.e. strategies) as you continue to feed it the latest market data. This also nearly eliminates datamining bias as you don't have a fixed data set on which you can overfit your ML, the data set is constantly changing.
Feb
2
comment How are risk management practices applied to ML/AI-based automated trading systems
@shabbychef I think online machine learning can mitigate those risks for the most part.
Feb
2
revised How can I go about applying machine learning algorithms to stock markets?
Clarification.
Feb
2
awarded  Teacher
Feb
2
answered How can I go about applying machine learning algorithms to stock markets?
Feb
1
comment What broker/feed/APIsetup allows for recording the most accurate data (cheaply)?
@allen, it should be pretty easy to check if you download the T4 desktop and open up one of the sample projects from the installation directory. I think you will also need Visual Studio 2008 or later.