SERVING THE QUANTITATIVE FINANCE COMMUNITY

ISayMoo
Topic Author
Posts: 2368
Joined: September 30th, 2015, 8:30 pm

Re: If you are bored with Deep Networks

Fair point about the PDEs. But there's more to ML than naive attempts at PDE solving.
True; but ML-PDE js the talk of the town in the quant sphere.
What about using NLP for sentiment analysis in quant trading? That's also ML.

ISayMoo
Topic Author
Posts: 2368
Joined: September 30th, 2015, 8:30 pm

Re: If you are bored with Deep Networks

PDE/ODE is an endangered species. And several MFE programmes demand such savoir faire for entry. And Python is easier to learn than C++ in this regard.
Do you need any programming language to learn PDE theory? It's maths.

Cuchulainn
Posts: 63368
Joined: July 16th, 2004, 7:38 am
Location: Amsterdam
Contact:

Re: If you are bored with Deep Networks

PDE/ODE is an endangered species. And several MFE programmes demand such savoir faire for entry. And Python is easier to learn than C++ in this regard.
Do you need any programming language to learn PDE theory? It's maths.
It's not necessary and probably not sufficient.

In solving optimisation problems in function spaces, Euler made extensive use of this method of finite differences'. By replacing smooth curves by polygonal lines, he reduced the problem of finding extrema of a function to the problem of finding extrema of a function of n variables, and then he obtained exact solutions by passing to the limit as n ! 1. In this sense, functions can be regarded as functions of infinitely many variables' (that is, the infinitely many values of x(t)at different points), and the calculus of variations can be regarded as the corresponding analog of differential calculus of functions of n real variables.

Cuchulainn
Posts: 63368
Joined: July 16th, 2004, 7:38 am
Location: Amsterdam
Contact:

Re: If you are bored with Deep Networks

Fair point about the PDEs. But there's more to ML than naive attempts at PDE solving.
True; but ML-PDE js the talk of the town in the quant sphere.
What about using NLP for sentiment analysis in quant trading? That's also ML.
Is it a black box? or more like HMM?

ISayMoo
Topic Author
Posts: 2368
Joined: September 30th, 2015, 8:30 pm

Re: If you are bored with Deep Networks

What is the definition of a "black box"?

Cuchulainn
Posts: 63368
Joined: July 16th, 2004, 7:38 am
Location: Amsterdam
Contact:

Re: If you are bored with Deep Networks

What is the definition of a "black box"?
I'll give it a go

1. One in which cannot interpret how input is processed to output
2. Given an output, how did we arrive at it (from primary and intermediate input).
3. (output is repeatable?) Black boxes that are not repeatable useful?(?)
4. The typical software system is a monolithic black box..

In software systems BB is well-established.
https://en.wikipedia.org/wiki/Black_box

ISayMoo
Topic Author
Posts: 2368
Joined: September 30th, 2015, 8:30 pm

Re: If you are bored with Deep Networks

1. Many ML models are relatively simple: construct a set of features and use them to train a linear or logistic regression "top model". They may be cumbersome to interpret due to the dimensionality of the data and feature space, but not difficult.

2. Is a question, not a requirement. What do you mean?

3. The output of a trained model is usually repeatable.

Cuchulainn
Posts: 63368
Joined: July 16th, 2004, 7:38 am
Location: Amsterdam
Contact:

Re: If you are bored with Deep Networks

I usually respond to specific points that I understand. The problem is, often your points are vague and hard to understand. And you don't seem open to learning things. At least that's my impression.
Seems I am not alone. And I am talking about PDE-ML.
The threads on Numerical are much better as they attempt at least to run to completion.

Many ML articles are preaching to the converted. Each to his own, but ML/CS does not understand PDE. Look here (JohnLeM is a expert PDE guy)

https://forum.wilmott.com/viewtopic.php?f=34&t=101716
And the authors never respond.

You can't say we didn't try.
"I am sure that, today, math is not necessary to machine learning and artificial intelligence. Indeed a mathematician can t understand what the hell they are doing However I am pretty convinced that this is only a matter of time !"

Cuchulainn
Posts: 63368
Joined: July 16th, 2004, 7:38 am
Location: Amsterdam
Contact:

Re: If you are bored with Deep Networks

I usually respond to specific points that I understand. The problem is, often your points are vague and hard to understand. And you don't seem open to learning things. At least that's my impression.
Seems I am not alone. And I am talking about PDE-ML.
The threads on Numerical are much better as they attempt at least to run to completion.

Many ML articles are preaching to the converted. Each to his own, but ML/CS does not understand PDE. Look here (JohnLeM is a expert PDE guy)

https://forum.wilmott.com/viewtopic.php?f=34&t=101716
And the authors never respond.

You can't say we didn't try.
Anonymous quote
"I am sure that, today, math is not necessary to machine learning and artificial intelligence. Indeed a mathematician can t understand what the hell they are doing However I am pretty convinced that this is only a matter of time !"

ISayMoo
Topic Author
Posts: 2368
Joined: September 30th, 2015, 8:30 pm

Re: If you are bored with Deep Networks

Completely disagree with it.

FaridMoussaoui
Posts: 507
Joined: June 20th, 2008, 10:05 am
Location: Genève, Genf, Ginevra, Geneva

Re: If you are bored with Deep Networks

anonymous quote --> Jean-Marc?

Cuchulainn
Posts: 63368
Joined: July 16th, 2004, 7:38 am
Location: Amsterdam
Contact:

Re: If you are bored with Deep Networks

Oui

ISayMoo
Topic Author
Posts: 2368
Joined: September 30th, 2015, 8:30 pm

Re: If you are bored with Deep Networks

Understanding ML requires a solid knowledge of mathematics. The theory behind the RBF kernels, for example, involves functional analysis and infinitely dimensional Hilbert spaces.

bearish
Posts: 5980
Joined: February 3rd, 2011, 2:19 pm

Re: If you are bored with Deep Networks

I once did something bad in an infinite dimensional Hilbert space. As a step in an alleged proof I asserted that the closed unit ball was compact. Good thing I took the class pass/fail...

katastrofa
Posts: 9702
Joined: August 16th, 2007, 5:36 am
Location: Alpha Centauri

Re: If you are bored with Deep Networks

Understanding ML requires a solid knowledge of mathematics. The theory behind the RBF kernels, for example, involves functional analysis and infinitely dimensional Hilbert spaces.
And that's just for Deep Learning. Can you tell how many concepts from different disciplines one must know to understand Reinforcement Learning?