Category Archives: Uncategorized

Machine Learning Math – The difference between A and B if A == B is that the history is different. A might come from sum(vec) = A. Different future? Pi and probability belongs a little since the decimals of pi show random distribution parameters // Per

February 22, 2019 Per Lindholm

Machine Learning Agriculture – Allow the plants to iterate?. Energy efficiency, (Th-Tc)/Th = n related to plant problems. Guess. If looks like the plant is enjoying themself maybe they are. Leaf shelter if too much sun == drapes?

February 22, 2019 Per Lindholm

Machine Learning Guess – Over Problemsolvning – Label simulations that are related to the specific problem. Instead of direct transformation learning. Then ?fit the parameters of the simulation so it becomes more relevant to the problem.

February 22, 2019 Per Lindholm

Machine Learning Math Guess – Invent any imaginary Problem and solve them and you get a more complete Mathematical Framework. For example. Math Forces are ?objects that move problems out of deep layer models()?

February 21, 2019 Per Lindholm

Machine Learning Physics – Black Hole Network. When in doubt assume a network. Gives a black hole is a network that stops harmful things coming out. Black holes ?are not new physics but new intelligence due a lot of added problems.

February 21, 2019 Per Lindholm

Machine Learning Math – Thought Math – Draw a bubble around a number or variable in an equation and imagine what this could mean. Then try to program solve a problem like this in machine learning.

February 21, 2019 Per Lindholm

Machine Learning Paradigm Thought – If You Cant Split The Problem In Part Truths You Can ?Always Split It Into Many Guesses // Per Lindholm 2019-jan-27 Guess Mathematics

February 21, 2019 Per Lindholm

Idea – Could a cause of a vulcano eruption be that not wanted material got under into the lava? CO2? So a the eruption has some meaning and is not just a physical effect.

February 21, 2019 Per Lindholm

Machine Learning Idea – Using GAN asking to solve an equation. Showing x,y,z as the iterate towards the answer // Per

February 21, 2019 Per Lindholm

Machine Learning Guess – As a possibility you can wait for some of equations in the system. Why solve all at the same time. Could produce more complex paths otherwise. Here I waited for y1-y2 -> 0

February 18, 2019 Per Lindholm

Machine Learning Innovation Idea – Generalisation by same language model0(). y1 = model1(model0(X1)), y2 = model2(model0(X2)),y3 = model3(model0(X3)),… Where X1 = MNIST sample and X2 = Cifar10 sample And X3 = something else in image format. Image showing most linear recognition //Per

February 16, 2019 Per Lindholm

Towards a Machine Learning Super Battery – Are different laser printed hole patterns different in their ion properties like capacity and transport? Laser aided CD printed like battery// Per

February 15, 2019 Per Lindholm

Machine Learning Innovation – Air Condition Device – If There is a need of a compressor. I wonder if you can alter the surface function with a cylinder. Or better yet use a peltier element connected to the front end of the piston. So it cools the hot gas. //Per

February 14, 2019 Per Lindholm

Blender 3D Innovation Idea – What if you take three render models() and iterate until you have enough intersecting pixel color values. Could this set be the inital value basis of a differential equation that solves the overall image with super high resolution? Venn Diagram Usage

February 14, 2019 Per Lindholm

Towards A Super Battery – Inspired by ?saltpaper. Could sandpaper or paper with crystals be a no brainer thing of a super battery?

February 13, 2019 Per Lindholm

Machine Learning Physics Idea – Predicting simulated turbulence with a w = model(rnge()).

February 13, 2019 Per Lindholm

Machine Learning Math Idea – choice_division(a,b). Where you get a rest that moves between the even parts. Dynamic object?

February 13, 2019 Per Lindholm

Machine Learning Physics – My most important tweet ever. Small fusion so that we have time and money to get the data samples. Million+?

February 10, 2019 Per Lindholm

Machine Learning Complex Math – Doing some complex root finding with machine learning model() and chainer framework. y**2 = 1+1j gives y = 1.0986842+0.45508987j wow

February 10, 2019 Per Lindholm

Machine Learning Physics – Force Or Acceleration Which Comes First. ( ? ) * F = m * a * ( ? ). My answer is that both are something else at the very start and that ( ? ) iterates fast to 1. // Per Lindholm 2019

February 9, 2019 Per Lindholm

Ultra Intelligent Way To Learn – First Read a little Bit Then Innovate By Guessing a Next Little Step. Read the next little Bit and guessinnovate the next. Keep doing so and you will be able to innovate forever.

February 9, 2019 Per Lindholm

Machine Learning 0..1 Math – Finding new functions? Math Universe needs to Split the first largest areas of Sum(every function the x = 0..1 window). Indication of a new function. Fill in the holes in the 0..1 window and you get something like a fractal. Conjectory – All functions applied to the window x = 0..1 makes up a fractal.

February 9, 2019 Per Lindholm

Fill in the holes in the 0..1 window and you get something like a fractal. Conjectory – All functions applied to the window x = 0..1 makes up a fractal // Per Lindholm 2019-feb-09

Machine Learning Insight Idea – Internally model() Initiated For Loops? For when there is no iteration-path and you have to use a little brute force.

February 9, 2019 Per Lindholm

Machine Learning Insight – Super Duper Calculations – y , sane = Classification(Regression(X)) , sane(y)

February 9, 2019 Per Lindholm

My Machine Learning Math Limit Idea – List Limits – Defined as lim(a,a->np.any(list)) where list = [n0,n1,n2,…n] // Per Lindholm 2019-feb-09 morning

February 9, 2019 Per Lindholm