Why is statistics so boring?

All serious intellectual endeavors are hard work. You could be playing video games or scrolling memes. If you want to build neural networks you need to learn statistics. Mean squared error, linear regression. It is hard work. It means spending hours focused on "boring" concepts so you can do interesting things. I think math is beautiful. It feels mystical in a way. I enjoy these things.

>>16705406

With statistics you can predict things.

You can predict words and build a neural network that talks like a human being.

You can predict images and draw a picture or a video out of a sentence of words.

You can predict where it will rain and when, the likelihood of alien life in the universe, just about anything.

Statistics are not boring when you think about it. The things you can do with statistics can appear as magic to most people.

Consider an evolutionary neural network (ENN), it is basically a graph of nodes, neurons, that do their best to reduce statistical error or loss. It can build itself by generating random data.

It is like a mutating and evolving intelligent lifeform. It generates a random move, makes a mistake, fails and tries again. Over and over, generations simulated with high powered hardware until it learns to reach the correct answer, to succeed at the task.

These self-building networks can "learn" on their own how to do anything. You give it a goal, a series of outputs to interact with the world, training data to learn from, and simulate it for generations of random trial and error. Yet it is made out of pure mathematics and logic, it is not a physical thing, it is a conceptual mathematical relationship simulated with 1s and 0s, just sets of numbers being multiplied and adjusted. It is made out of pure math, namely statistics. These are at the cutting edge of science and are being published in recent decades.

>>16705420

Using stastics, we can build an artificial intelligence that is more intelligent than human beings and more efficient than learning by many magnitudes. I do not think it is boring at all. Alan Perlis, computer scientist, says that a year spent in artificial intelligence is enough to make one believe in God.

>>16705422

>more efficient at learning by many magnitudes

Computers of course are already better than us at basic arithmetic and playing chess, but the difficulty is in teaching them to think like a human being. This requires the multiplication of matrices, it is more than flipping 1s and 0s but it is ultimately reduced to flipping 1s and 0s.

It is mind-boggling to consider the calculus of higher dimensions. We have to reduce it to these nodes because it goes well beyond 2D and 3D conceptualization, two or three dimensions. We are talking thousands of dimensions and calculating the derivstions of a "slice" of each dimension one at a time to make tiny adjustments. It requires immense computing power. GPUs are best suited for it, as are supercomputers. A desktop CPU will take hours to run a model that can predict images only about 100 pixels wide.

>>16705424

>We are talking thousands of dimensions and calculating the derivstions of a "slice" of each dimension one at a time to make tiny adjustments.

Past three dimensions my mind can't conceptualize it. I can visualize the derivation of a 3D graph making steps to find local minima through linear regression. It is like a ball falling to a curve in a surface.

4D? What does that look like? 100+ dimensions? Just to draw a tiny image. I can only visualize the nodes. At this point the computer is thinking on a level the human brain just isn't capable of.

>>16705427

All of this is statistics. It is not boring at all. When you are doing linear regression in higher dimensions it is the realm of science and research. It is absolutely essential to all hard sciences. Calculating error and loss.

>>16705427

>>We are talking thousands of dimensions and calculating the derivations* of a "slice" of each dimension one at a time to make tiny adjustments.

These models have use in all sciences. They can be used to solve problems and predict things. Probably one of the most classic applications of linear regression is modeling population growth in only two dimensions. Important research that gives useful insights.

>>16705375 (OP)
Would an alien actually do this.

>>16705427
>Past three dimensions my mind can't conceptualize it
>What does that look like? 100+ dimensions?
It's simple really, first you imagine n-dimensional space, then you set n to 101.

>>16705422
>we can build an artificial intelligence that is more intelligent than human beings and more efficient than learning by many magnitudes.

>>16705427
What type of 4 dimensional data do you look at?

>>16705375 (OP)
It is boring. Just because it has intresting applications does not make it fun like the other tards in this thread claim.

Probability thoery and stochastic processes are fun but statistics is boring because you have to spend time looking at errors and p values like some science tard.

>>16705375 (OP)
Because a lot of it is approximations and assumptions. And then the art of it is calculating things around that.
>muh hecking curve fitting
>muh hecking p values
>muh hecking curtosis
It’s interesting to study distributions on their own until you realize that 99% of use cases don’t actually admit known distributions and we’re just extracting a bunch of numbers from a data set.

>>16705427
>4D? What does that look like?
go play golf
https://www.youtube.com/watch?v=yQOFzsRXG1Y

>>16705422
>jewish academic says his research is literally God
Financial incentives and a genetic predisposition to schizophrenia

>>16705422
We all know my mind looks nothing like this. I need to avoid this thread like the plague but I'm not going to. One of you has to have a rough idea how my mind works. "You're just incredible math." Isn't good enough.