Monthly Archives: March 2018

Linux Idea – Use A Jupyter Notebook For Hard To Remember Terminal Commands?

The idea is to use a Jupyter notebook in a handy way. Be sure to use functions as the commands otherwise you might run them all in one batch. And dont delete your files.

It could be the start of a Smart Terminal. For example. You can create a function safe(). That takes the command and outputs warnings before execution or simple filters bad beginner mistakes.

 

Philosophy – What Is A Power Sentence And Why Is It Important?

A power sentence is simply a bold statement that when true will give you a much deeper understanding and a better knowledge. It will generate innovative new guesses and much more.

My best power sentence is the general answer to question why.

– What Ever The Question Might Be The Answer Must Make Sense To A Network.

With this power sentence you can answer previously thought impossible questions.

CPU Cooling Idea – Air Suction As A Way To Remove The Heat From The CPU

The idea is simple.

My only invention home with suction is the vacuum cleaner. So from this. If we were to invent efficient suction. What other inventions could we come up with that uses this method of mass transport?

Like for the CPU. What if there we used the suction method instead of blowing air through metal sheets. I’m thinking. What if the CPU cooler were a block with tiny hole tubes in it. Like you need to draw air through a straw.

Homeless Prevention – Pensioner Backup Housing?

I wondering. Should there exist a goverment plan for backup housing for pensioners? To prevent a lot of people becoming homeless. Specially when your older and you lack the strength.

Could not pensioner funds be used for something like this. Pensioners taking shared responsibility for each other in building cheaper housing.

Be it tiny houses with tiny rent, affordables or something.

Machine Learning Math – Example Mathematics Train Models On Graphs And Labels

A fast machine learning idea. So I can party like its 999+ posts.

The idea goes like follows. Plot some mathematical concept like the mean value. Portray it by a graph of points and possibly some line for the mean value.

Train several of these graphs images and corresponding labels on the machine learning model.

Voilà. You have a crude mathematical machine learning model.

Math Idea – Should There Exist Intensity Equations Complementing Differential Equations?

I guess differential equations are pretty bad at solutions that look like noise. So why beat the bush and instead define a new category of methods. Call them Intensity Equations or something similar.

So then something like violent turbulence could be ?solved by formulating an intensity equation.

Maybe the solutions is not possible to get by hand since the amount of information in the information dense picture of the real world sample is pretty large.

So the intensity equation should ?with the help of a generator, G(noise) create an information dense solution.

Machine Learning Innovation – How Would Non-Linear Connections Work?

Playing around with a new machine learning library. Neural Networks in Python, NeuPy. I came across an example of connection visualization.

The connections does not seems all that complex. Hmm here might be room for complex improvement. Playing around with ’heal section” under GIMP I created a non linear mock image of the connections. Maybe this can serve as inspiration.

So how can non-linear connections work and improve machine learning?

Notice the healed section of non linear looking connections.

Innovate Bangladesh – Using Machine Learning To Create An Open Vectorized Video Codec For Low Bandwidth Youtube?

The idea is simple.

Vectorized video I think can reduce the size of a video more than 250 times. Thats a very good. So good that I think its worth creating a youtube javascript viewer for that would be an open format.

This way people with ?250 times slower broadband can watch and enrich their lives. Create businesses around youtube subscribers like never before.

I believe machine learning is the key to create such a decoder/encoder.

Improve Healthcare Research Idea – Could Machine Learning Help Earn You More Money? Store Layouts etc.

If you want to improve healthcare. Find out ways with or without using machine learning how to improve peoples lives. More money more and better healthcare.

So for a store owner. Could an imaginative GAN machine learning network based on a set of successful previous store layouts improve your store layout. The placement of food, products etc.

Machine learning is in the smart imagination not necessarily in big data.

Innovate Bangladesh – Carpenters For Borrow? Public Funded Public Service

To create a win-win for refugees and host countries. Now for the Rohingyas. I thought. Why not come up with business plans to increase the wealth of Bangladesh.

So I thought the ?easiest way would be to empower the poor. First I thought. Could a library of tools work. But the tools would probably be lost.

This gave me another idea however. What about tool equipped carpenters for borrow. The idea is that you can borrow a carpenter to build whatever the people need to build to increase their sales.

Perhaps a stand on wheels to sell vegetables could be something to build.

Better broadband access would be another item to create.

It could be that carpenters for borrow is so successful that we also can have that as a public funded public service. Building tiny houses and more.

Machine Learning Idea – Should There Exist An Index Generator Or An Intelligent Sampler?

I have an idea that because there is no a single right way to convert a 2D matrix to a long vector I would assume its a complex process.

That is. I think its possible to take advantage of the model optimizing capability of machine learning libraries to generate an index vector as a conversion tool.

The length of the index select vector need not be of exact length of the data. It could be smaller, exact or bigger. I guess it could sample problem areas much more than others. Generating duplicates of data in the same long vector.

So whenever there is a problem assume a network model. Like which data come clustered together in the vector.

Dyslexia Idea – Inspiration From Machine Learning. Three Shifted Rows With The Same Sentence. Large In Data Matrix.

I experienced some visual loss yesterday. So remembering the blurred effect in the center of my focus. I tried to read but it was hard. Some focused words disappeared in a non focus recognition way.

Inspired by this and machine learning. One idea for those who have reading problems like dyslexic would be to increase the probability the target word the sentence sequence is recognizable.

So the idea is simply to duplicate or triple the same sentence in three rows under each other. Possibly some shifting.

This is not exactly as making the font larger though.

Environmental Idea – Keeping The Coral Reef Fresh As ?Easy As Keeping The Water In The Aquarium Clean. Optimize The Water For Aquarifish?

Plants and aquarifish need clean water to live. So does the coral reef. So the idea is to ?filter water for contaminants using solar power and some reverse osmosis like process. Continuously reinsert cleaned water to the coral reef surrounding to battle the speed of contamination.

Perhaps contamination could be slowed down by some smart enclosure of some size? But this is more complex I guess. Maybe some algorithmic enclosure where you have a machine learned algorithm to control water in and out transfer could help.

Machine Learning Idea – Imaginative Problem Probability Training Using GAN’s?

You can’t have a real world samples for every problem that could arise. But I think I have a start of a ?solution for this.

Using Generative Adversarial Networks or GAN’s. The kind of network that is trained to tell the difference between fake and real and generate real looking fakes.

It should be possible to create ’photo realistic’ fake problems. Then train a ?separate model on these imagination examples.

This way you extend real world example problem model with a imaginative based model which would be trained by a ?large number of real looking problems.

With a GAN network it would also be possible for a human engineer to do paint assist problems. That is. You would roughly paint the start of a problem and the network would iterate a photo realistic video of the problem.

One usage area ?could be self driving cars.

// Per Lindholm

Physics Speculation – Does There Exist Confusion Matrix Fields? Responsible For Randomizing Universe Object Properties Like Spin?

The ?spin directions of a material is suppose to go in random directions otherwise you get a magnet. The power guess here is that there is no such thing as a free lunch. Meaning that the level of randomness is achieved rather than given by nothing.

So the idea is that there could be a field, a confusion matrix field. Responsible for randomizing the electron spin directions. Other similar confusion fields would then be responsible for other universe objects which could carry random properties.

Education And Job Idea – What Jobs Can You Do To Save Us? That Will Be The Jobs People Are Willing To Pay A Salary For

So from this I’m convinced water and food are top priorities.

We need desalination engineers to reduce the cost of reverse osmosis. Perhaps by increasing complexity but reducing size with algorithmic means. Water engineer.

We need to restore farmland much quicker. Greening the desert is very important. Food engineer.

People will have less money. Meaning that the poor will have even less resources. We need ways to increase the little wealth of the poor. Perhaps through housing projects. Poor/Wealth Engineer

So we have many jobs people can do to save life on this planet.

Desert Engineer

Energy Engineer

Food Engineer

Water Engineer

Peace Engineer

Electric Car Engineer

Military engineer

Refugee engineer

Super Idea – How Much Food and CO2 uptake Do We Get With A Solar Powered Desalination Plant For The African Sahara Desert?

Perhaps its not so difficult to green the desert. To prevent hurricanes. To cut billions of losses due hurricanes. To make the whole desert green. What would the uptake of CO2 be alone. The Sahara Desert needs ??40+ desalinations plants only. With ongoing advances and investments in reverse osmosis desalination and solar power. It could be the best investment we can make. Apart from another 1000 other things we need to do ..

Innovate The Sahara Desert