Author Topic: AI - Artificial Intelligence  (Read 5425 times)

Jurgis

  • Hero Member
  • *****
  • Posts: 4072
    • Porfolio
AI - Artificial Intelligence
« on: May 17, 2018, 11:37:14 AM »
I did not find generic AI thread, so I'm gonna start one.  8)

In particular, I wanted to post a link to this: https://www.msn.com/en-us/news/other/the-world-e2-80-99s-dominant-crypto-mining-company-wants-to-own-ai/ar-AAxqyH5 (article connected to cryptocurrencies, NVDA, AI ASICs).
« Last Edit: June 08, 2018, 06:32:22 AM by Jurgis »
"Before you can be rich, you must be poor." - Nef Anyo
--------------------------------------------------------------------
"American History X", "Milk", "The Insider", "Dirty Money"


Liberty

  • Lifetime Member
  • Hero Member
  • *****
  • Posts: 10369
  • twitter.com/libertyRPF
    • twitter.com/libertyRPF
Re: AI thread
« Reply #1 on: May 17, 2018, 11:48:25 AM »
This video showing a simple (and entertaining) explanation of how machine learning probably belongs here, then:

https://www.youtube.com/watch?v=R9OHn5ZF4Uo
"Most haystacks don't even have a needle." |  I'm on Twitter  | Interesting podcast on aging research

Jurgis

  • Hero Member
  • *****
  • Posts: 4072
    • Porfolio
Re: AI thread
« Reply #2 on: June 07, 2018, 04:31:21 PM »
Looks like the name of the thread was not great. Can't find it using search function.  :( Should have named AI - Artificial Intelligence.  ::) I don't think I can rename right now.

Anyway, today I've read pretty impressive paper on predicting index X using deep neural nets using 20XX-20XX dataset. Great results. Not sure if this is one of these situations where back testing works and currently the approach stopped working or not. I would have thought that their approach should not have worked on historical data either, since this seems should have been exploited already. They do have training set and test set and some additional checks, so unless they did something "bad", the results seem to be real.

Edit: After some thinking, I have some questions/ideas/reservations/inquisitions to test if this really works or not. If anyone is familiar with DNNs and want to run some experiments, shoot me a message and we can play around. Results could be contribution to science or monetary (if you get a model that works, you can just use it to invest  ;D - most likely you won't ). I might do it myself, but I need time and motivation... yeah, yuge pile of money is not motivation enough.  :P  ::)  8)  ;D

I'll point to the paper when it becomes publicly available.  8)
« Last Edit: June 07, 2018, 04:46:00 PM by Jurgis »
"Before you can be rich, you must be poor." - Nef Anyo
--------------------------------------------------------------------
"American History X", "Milk", "The Insider", "Dirty Money"

Liberty

  • Lifetime Member
  • Hero Member
  • *****
  • Posts: 10369
  • twitter.com/libertyRPF
    • twitter.com/libertyRPF
Re: AI thread
« Reply #3 on: June 08, 2018, 06:16:59 AM »
Looks like the name of the thread was not great. Can't find it using search function.  :( Should have named AI - Artificial Intelligence.  ::) I don't think I can rename right now.


If you hit "modify" on the original post in the thread, you can edit the title.
"Most haystacks don't even have a needle." |  I'm on Twitter  | Interesting podcast on aging research

Jurgis

  • Hero Member
  • *****
  • Posts: 4072
    • Porfolio
Re: AI - Artificial Intelligence
« Reply #4 on: June 08, 2018, 06:32:56 AM »
Thanks. Done.
"Before you can be rich, you must be poor." - Nef Anyo
--------------------------------------------------------------------
"American History X", "Milk", "The Insider", "Dirty Money"

cameronfen

  • Full Member
  • ***
  • Posts: 234
Re: AI thread
« Reply #5 on: June 12, 2018, 04:37:47 AM »
Looks like the name of the thread was not great. Can't find it using search function.  :( Should have named AI - Artificial Intelligence.  ::) I don't think I can rename right now.

Anyway, today I've read pretty impressive paper on predicting index X using deep neural nets using 20XX-20XX dataset. Great results. Not sure if this is one of these situations where back testing works and currently the approach stopped working or not. I would have thought that their approach should not have worked on historical data either, since this seems should have been exploited already. They do have training set and test set and some additional checks, so unless they did something "bad", the results seem to be real.

Edit: After some thinking, I have some questions/ideas/reservations/inquisitions to test if this really works or not. If anyone is familiar with DNNs and want to run some experiments, shoot me a message and we can play around. Results could be contribution to science or monetary (if you get a model that works, you can just use it to invest  ;D - most likely you won't ). I might do it myself, but I need time and motivation... yeah, yuge pile of money is not motivation enough.  :P  ::)  8)  ;D

I'll point to the paper when it becomes publicly available.  8)

I have decent amount of expeirience with deep neural nets but mostly in LSTMs and not conv nets.  My friend has a ai trading strat that is incredibly successful, but I told him if I worked,in the area I'd find something else.  PM me if interested in discussing more. 

winjitsu

  • Full Member
  • ***
  • Posts: 147
Re: AI thread
« Reply #6 on: June 13, 2018, 05:57:19 PM »
Looks like the name of the thread was not great. Can't find it using search function.  :( Should have named AI - Artificial Intelligence.  ::) I don't think I can rename right now.

Anyway, today I've read pretty impressive paper on predicting index X using deep neural nets using 20XX-20XX dataset. Great results. Not sure if this is one of these situations where back testing works and currently the approach stopped working or not. I would have thought that their approach should not have worked on historical data either, since this seems should have been exploited already. They do have training set and test set and some additional checks, so unless they did something "bad", the results seem to be real.

Edit: After some thinking, I have some questions/ideas/reservations/inquisitions to test if this really works or not. If anyone is familiar with DNNs and want to run some experiments, shoot me a message and we can play around. Results could be contribution to science or monetary (if you get a model that works, you can just use it to invest  ;D - most likely you won't ). I might do it myself, but I need time and motivation... yeah, yuge pile of money is not motivation enough.  :P  ::)  8)  ;D

I'll point to the paper when it becomes publicly available.  8)

Just cross validation during bull market years? I've played around with it a bit but never been comfortable enough with the algo [even worst with NNs]. I'm very scared of blowing up with these over-fitted models that have only seen rising markets...

I think the main criticism against these "paper" strategies is you have 1000s of academics looking for signals and the winners publish a paper. The signals they find basically are the result of survivor bias.

Do you guys have slack? Maybe its time we start a CoBF slack group 

Jurgis

  • Hero Member
  • *****
  • Posts: 4072
    • Porfolio
Re: AI thread
« Reply #7 on: June 14, 2018, 08:57:42 AM »
Just cross validation during bull market years?

No. First, you can't - or you shouldn't - cross validate in conventional sense, since you should not apply algo trained on the future data to the past data. At least these guys don't do it and IMO that's a correct approach. I guess you can do limited cross validation with that restriction, but it limits your data amount possibly a lot, which means worse training and worse results I'd guess (see below too).

They do something clever, but they don't explain it well, so it may be "really smart" (TM) or "really not helpful" (TM).

Quote
I've played around with it a bit but never been comfortable enough with the algo [even worst with NNs]. I'm very scared of blowing up with these over-fitted models that have only seen rising markets...

Well, first I assume you already split data into training/dev/test set, yes? So you should not overfit, otherwise you won't get good results on test data.
Second, why you assume that you train/test on only rising markets? There's data for more than that... you can test on data that includes 2007-2009 crash, no? :)

Though I agree that there are issues:
1. Depending on what you are training/etc. there might be not-much data. E.g. if you train on daily prices, and you have 10 years of data, that's ~3.5K data points, which is quite low when you think about NN training sets.
2. Even if you include times with crashes, there might be only 1-2 big crashes per training/test data, which is also quite sparse in terms of data... so yeah, there's a risk that a "different" downturn/crash/whatever may not be handled (well).

Quote
I think the main criticism against these "paper" strategies is you have 1000s of academics looking for signals and the winners publish a paper. The signals they find basically are the result of survivor bias.

I don't think this is the case, but as I said, I have other reservations.

Quote
Do you guys have slack? Maybe its time we start a CoBF slack group

No, I don't have slack. 8) But if you guys want to move further discussion to a limited group, we can organize something.  8)
"Before you can be rich, you must be poor." - Nef Anyo
--------------------------------------------------------------------
"American History X", "Milk", "The Insider", "Dirty Money"

TheAiGuy

  • Sr. Member
  • ****
  • Posts: 324
Re: AI - Artificial Intelligence
« Reply #8 on: June 14, 2018, 09:58:05 AM »
Wow, so good luck with this? This should be a really, really hard thing to do without a good dataset. I imagine most quant funds (e.g. Renaissance or Bridgewater) have access to large, high quality datasets. I seriously doubt you will do well running LSTMs on widely available datasets (also, donít use LSTMs - use GRUs)

Anyway, if you are serious about this, a good place to start for tools is probably Quantopian. I know one of the principals there and I donít think I can vouch for their financial market chops but there toolsets are probably pretty good (i.e. their python interfaces)

TheAiGuy

  • Sr. Member
  • ****
  • Posts: 324
Re: AI - Artificial Intelligence
« Reply #9 on: June 14, 2018, 09:58:37 AM »