What to feed to the input of the neural network? Your ideas... - page 43

New comment
 
Aleksandr Slavskii #:

I apologise for the off-topic.

I came across a saying:

If you get shit on, it means you made someone shit on you.


I'm talking to those who shit themselves. Get out of here, you stink of shit.

I join you in wishing you well.
 

Denial, bargaining , anger, depression, acceptance.

The most interesting phase is coming up

 

"What do I feed into the neural network?"

Whatever is fed in, it's all past the money.) And, actually, the sequence of quotes, without the nonsense.

 
Andrey Dik #:

Discrete AO.)) There is no such notion, all AOs are discrete, discreteness is defined by the step of optimised parameters.

https://en.wikipedia.org/wiki/Discrete_optimization



All AOs are divided into discrete optimisation and continuous optimisation, this is the very first branch in AO subspecies, that is the most elementary ....

It's like writing articles about MO but not knowing that it is divided into classification, regression and clustering, etc...

You are not just an ignoramus, you are the king of ignoramuses, you don't know anything about optimisation at all.



And on the second claim it is just a face palm....

 
mytarmailS #:

https://en.wikipedia.org/wiki/Discrete_optimization

All AOs are divided into discrete optimisation and continuous optimisation, this is the very first branching in AO subspecies, i.e. the most elementary ...

It is like writing articles about MO but not knowing that it is divided into classification, regression and clustering, etc....

You are not just a profane, you are the king of profanes, you don't know anything about AO at all.

You can bitch as much as you want from impotent anger, it is not clear on whom and for what, you will be taught by wiki and you will remain what you are, a militant ignorance. Ask your friend about how to write a wiki, he will tell you. And think with your head. There are no continuous AOs, it's not a wave process, they are all discrete with a finite step size.

 

Prompt:

What are the sections of optimisation

Response:

## Sections of Optimisation


Optimisation is a vast field covering various tasks and methods. Dividing optimisation into sections helps to better understand its structure and find the right tools to solve specific problems. Here are some of the main sections of optimisation:

**1. By type of problem:**

* **Linear Optimisation:** deals with problems where the target function and constraints are expressed by linear equations. It is one of the most studied sections of optimisation, with many algorithms and software tools.
* **Nonlinear Optimisation:** deals with problems with non-linear functions, which complicates the search for an optimal solution. Various methods are used in this section, including gradient methods, Newton's methods, and evolutionary algorithms.
* **Combinatorial Optimisation:** focuses on problems involving the selection of the best option from a finite set. Classic examples are the destination, route, and backpacker problems.
* **Stochastic Optimisation:** deals with problems with uncertainty where the input data or parameters may be random variables. Methods of probability theory and statistics are used to find an optimal solution.

**2. By field of application:**

* **Optimisation in economics:** is used to solve problems of resource management, production planning, pricing, etc.
* **Optimisation in engineering:** used for design of structures, process control, optimisation of control system parameters.
* **Optimisation in medicine:** helps in solving problems of diagnosis, treatment, planning of clinical trials.
* **Optimisation in logistics:** used to optimise delivery routes, warehouse management, transportation planning.

**3. By solution methods:**

* **Mathematical programming:** uses methods of mathematics such as linear algebra, differential calculus, graph theory to construct and solve optimisation models.
* **Heuristic methods:** rely on empirical knowledge and rules of thumb to find a "good enough" solution without guaranteeing its optimality.
* **Artificial intelligence methods:** apply techniques such as machine learning, neural networks, evolutionary algorithms to solve complex optimisation problems.

**4. By type of variables:* *

* **Optimisation with continuous variables:** operates with variables that can take any value within a given range.
* **Optimisation with discrete variables:** deals with variables that can take only certain, predetermined values.
* **Optimisation with Boolean variables:** deals with problems where variables can take only two values, "true" or "false".

**5. By type of constraints:**

* **Optimisation with constraints:** deals with problems where there are constraints on the allowed values of variables. These constraints can be represented as equations, inequalities or logical conditions.
* **Optimisation without constraints:** deals with problems where there are no constraints on the values of variables.

**6. By type of target function:**

* **Minimisation:** the objective is to find a value of the variables that minimises the value of the target function.
* ** **Maximisation:** the goal is to find the value of the variables that maximises the value of the target function.

This is by no means a complete list of the sections of optimisation. Each of them has its own methods, algorithms and applications. The choice of a particular section and method of optimisation depends on the specifics of the problem to be solved.

## Additional materials:


 
Yuriy Asaulenko #:

"What to feed the input to the neural network?"

Whatever is fed in, it's all past the money.) And, actually, the sequence of quotes, without the nonsense.

How many years, how many winters! A warm welcome!

 
mytarmailS #:

https://en.wikipedia.org/wiki/Discrete_optimization

All AOs are divided into discrete optimisation and continuous optimisation, this is the very first branching in AO subspecies, i.e. the most elementary ...

It is like writing articles about MO but not knowing that it is divided into classification, regression and clustering, etc....

You are not just an ignoramus, you are the king of ignoramuses, you know nothing about optimisation at all.

And the second claim is just a face palm....

Even before I saw it on the forum, I had time to write that so-and-so is completely cuckoo. I say oh, come on, with whom it does not happen.

...

 
Maxim Dmitrievsky #:

Even before I saw it on the forum, I had time to write that so-and-so is completely cuckoo. I said, "Oh, come on, it doesn't happen to anyone.

...

You're being humourous, but be careful.
 
Andrey Dik #:
You're being humourous, but you're being humourous.

It's factual.

1...363738394041424344454647484950...67
New comment