SI
SI
discoversearch

We've detected that you're using an ad content blocking browser plug-in or feature. Ads provide a critical source of revenue to the continued operation of Silicon Investor.  We ask that you disable ad blocking while on Silicon Investor in the best interests of our community.  If you are not using an ad blocker but are still receiving this message, make sure your browser's tracking protection is set to the 'standard' level.
Strategies & Market Trends : NeuroStock
 Public ReplyPrvt ReplyMark as Last ReadFilePrevious 10Next 10PreviousNext  
To: Jay Hartzok who wrote (764)1/17/2000 12:07:00 PM
From: Simon Thornington  Read Replies (2) of 805
 
Well my NS40S licence came in the mail this morning. I set it going on my XLNX-AMAT-COMP, but with 40 neurons. Something I'm concerned about, it seems to me (with very brief experience) that adding neurons might be a bad idea. With too many neurons, the net will do better by training itself too closely against the historical data, memory becoming more powerful than pattern detection. Has anyone experienced this? I wouldn't mind reading some books, there must be formula to determine the power of a certain number of neurons to analyze/remember certain amounts of data. (And then make sure that the net doesn't have enough neurons to remember the data too closely)

Of course, I don't know any of the math yet. If anyone has a book they could recommend, I'd appreciate it.

Cheers,

Simon.
Report TOU ViolationShare This Post
 Public ReplyPrvt ReplyMark as Last ReadFilePrevious 10Next 10PreviousNext  

HomeMailHotSubjectMarksPeopleMarksKeepersSettings
Terms Of UseContact UsCopyright/IP PolicyPrivacy PolicyAbout UsFAQAdvertise on SI
© 2025 Knight Sac Media.  Data provided by Twelve Data, Alpha Vantage, and CityFALCON News