Peroglyfer – Machine Learning changes math. Math changes everything.

Search
Skip to content
  • Home
  • List It Like 2039+ Ideas
  • My best ideas
  • About
  • Unsorted coolab notebooks
  • Changed Ideas

Monthly Archives: December 2018

Uncategorized

Refugee Education/Livelihood – With Internet Access They Could Become Fluent In Machine Learning With Services Like Google Colab. Run From Any Device With A Browser And Internet Connection. Upgrade To The Raspberry Pi From Slow Scratch Visual Code To Python GPU 11GB Accelerated

December 30, 2018 Per Lindholm
Uncategorized

Guess – If something is very ?unsure in nature like the position of the electron around the atom. Then it could be a sign that the position can be changed back in time. Since it would not matter to much to determinism. Adapting model() vs algorithm()

December 30, 2018 Per Lindholm
Uncategorized

Machine Learning idea – That take any object. The atoms that its made of are what the model() needs to generate. No more no less. To give the exact target answer. Like the smell, taste or shape. So the focusing should be on models that can generate the right lego objects?

December 30, 2018 Per Lindholm
Uncategorized

Energy Innovation – Fixed Price/Month From Livelihood Businesses – ?Let Small Entrprenours Sell Electricity From Their Own Wind and Solar Energy For A Fixed Price per month to electrify the rural country. The Business approach.

December 29, 2018 Per Lindholm
Uncategorized

Enable Climate Change Innovations Idea – Git-CAD 3D Models With Version control? So You can records changes to a file or set of files over time and recall specific versions later // Per Lindholm 2018-dec-29

December 29, 2018 Per Lindholm
Uncategorized

Idea – If Jupyter notebook Python had matplotlib, numpy and some better vesion of pip built in it would make things much easier. pip fails all to often. Otherwise we will become too much dependent on cloud infrastructure like colab with everything built in

December 29, 2018 Per Lindholm
Uncategorized

Linux Idea – Make Your Own “terminal” command in jupyter notebook. Like storing maximum iowait for each process. To see what slows the computer the most

December 29, 2018 Per Lindholm
Uncategorized

Rohingya And Bangladesh Innovation – Internal Job board. List of things that needs to be done or could benefit the refugee economy. Volunteer or otherwise compensated work?

December 29, 2018 Per Lindholm
Uncategorized

Rohingya Refugee Idea – Could an elected group of Rohingya men and woman be sponsored ambasadors for their people. Gathering support globally for peace, livelihoods, trade, business, education?

December 29, 2018 Per Lindholm
Uncategorized

Deep Dark Energy – Machine Learning Physics – Do the universe produce more energy for every calculation than necessary? Like my MNIST model(). y,res = F.separate(model(X).reshape(-1,2,10),axis=1). Where y is goes to the target and res are just some unused values

December 26, 2018 Per Lindholm
Uncategorized

Python Innovation – sensehat_doodles. As a start of learning a first language. Just do any kind of creative loop plot on the sensehat. Perhaps sensehat_doodles could exist as a standalone app on the mobile to import. Sensehat mobile?

December 26, 2018 Per Lindholm
Uncategorized

Battery Idea – Could the natural path experience of materials be used in a battery. Lithium found in the ocean should indicate that a high concentration to a very small concentration like in the ocean is useful information

December 25, 2018 Per Lindholm
Uncategorized

Solar Innovation – Set Of possibilities would suggest that there is a invention to be made from compressed CO2 surrounding a solar panel. Higher Efficiency? More solar since the sun wants a target CO2 concentration. Meaning dispersion of the CO2. The Sun is life optimized.

December 25, 2018 Per Lindholm
Uncategorized

Russia Innovation ? – Tiny Home Backup – What if you have system where people who wants to invest in risky? apartments get a Tiny Home On Wheels in addition. After the apartments are finished they can be returned to the housing company or serve as summer vacation houses.

December 25, 2018 Per Lindholm
Uncategorized

My Machine Learning Physics – ?Infinite reach gravity and an infinite number of stars means an infinite number of possibilities for universe objects. From that gravity could be a machine learning model(). Models() can do amazing stuff // Per Lindholm

December 24, 2018 Per Lindholm
Uncategorized

Migrant Innovation – How Many Dual Country Entrepreneurs Would It Take To Rebuild Venezuela. The idea is to Live abroad and Rebuild the Home Country. The faster the rebuild gets going the better for all.

December 24, 2018 Per Lindholm
There is always a way for challenges
Uncategorized

Machine Learning Physics – A Perfect Start Gives A Better Total Result – Longer Initial Iteration With The Same Input. Then Should Before Every Minimal Movement Of Mass Be Initial Iterations. Non or Minimal Moving Iterations. Perhaps Heat Should Adjust Its Information Pattern

December 23, 2018 Per Lindholm
Uncategorized

Any Mobile Phone Case Innovation Idea – Clip On Screen + Raspberry Pi Zero (Linux). For when termux is not enough.

December 22, 2018 Per Lindholm
Uncategorized

Machine Learning Gravity Guess – What If The Potential Energy Of An Object Is Encoded In The Earth Like A Weight Matrix. So When The Object Goes Higher. The Matrix Changes. This means the matrix weight path is not the same when an object goes down again. Its Not Reverting

December 22, 2018 Per Lindholm
Uncategorized

Refugee Digital Livelihood Idea – Turn Text To Speech Synthesis To Improve English For Non English Speaking People. That is. Real time perfection of the voice. English To Better Sounding English. This way people can take on more jobs

December 22, 2018 Per Lindholm

This idea could make billions. Improving english will improve business.

Uncategorized

Teeth Innovation – Just dip the toothbrush in mouthwash and brush. Its much safer and ?effective

December 21, 2018 Per Lindholm
Uncategorized

Machine Learning Physics Guess – Problem With Classification Of Dark Matter. What if you have both well Ordered mass and Noisy mass. This would be equal to the noisy weight matrices in machine learning and some denoised weight matrix. The universe could adapt to using both types // Per Lindholm 2018-dec-21

December 21, 2018 Per Lindholm

Why dark matter is a problem. Its a noisy classficiation?

Uncategorized

Homeless Innovation Idea – Buy Winter Homeless kits online to give to homeless this winter.

December 18, 2018 Per Lindholm
Uncategorized

Website Business or Product Investment Idea – Bitcoin Investments In Local Businesses To Rebuild Countries Affected By Climate Change And Migration

December 16, 2018 Per Lindholm
Uncategorized

Machine Learning Physics Guess – If Light Is A model(). Is speed the same as iteration through space? Meaning it needs n number of iterations per meter traveled?

December 16, 2018 Per Lindholm

Posts navigation

1 2 3 Next →

Ideas Are Free To Use As Long As You Give Me Credit

By Far The Smartest Idea - Hot Milk With Cooked Rice In A Cup

If free will can be recognized in the machine learning sense then it could probably also be optimized with max as the target. // Per Lindholm

Is a proof some knowledge of the probability of the network.

The Universe Is A Compression Decompression Model() // Per Lindholm 2018

Peace Idea - Any Type Of Probability Can Move Us Forward. Peace Will Do It 100% // Per Lindholm 2018

:wow

Peroglyfer Feed RSS
www.facebook.com/peroglyfer

Recent Posts

  • Breakthrough Machine Learning – Generalizing loss. Transform the loss 0..1 with a model2 the loss of model1. So if the loss oscilates wildy the generalizing loss dont. Its confined between around 0..1. Can the iterate with the same input for 100000 times without degrading. IDea and method by Per Lindholm
  • The Halting Problem – You can also learn if the function is going to converge or not. By random choice selecting between at=loss for converge = 1 and at= np.random()*loss for converge = 0 Then a model2 learns the converge label. Voila I guess the Halting problem is solved?
  • Machine Learning Idea – There are alot of zeros in machine learning target. So what if you can make patterns in the zeros. Assume some kind of zeroToSmallValue Pattern is beneficial to the classification. // Per
  • Machine Learning – Generalizing Idea – Sometimes convolutions produce shadows. Shadows from different objects can appear the same. I think you also get generalization from ||derivative similarity||. With a some small number of value similarities. // Per
  • Safe Affordable Life – Innovation – Wash Your Clothes Securely at the workplace. #Africa #Sweden #US
  • Electric Airplane Energy Strategy – Since electric motors and battery are good at acceleration. Could a strategy be to accelerate faster the plane for longer periods of time. Accelerate – glide – accelerate? for electric Cargo Planes?
  • ?Lindholm Convolution – Inspired by random shape images. I wondered what x belong to an image which pixels colors are assumed to be random. cumsum(x-mean(x)) = image. I think I found a convolution style. // Per
  • Association Intelligence – Random numbers has something to do with shapes and figures. Since you sometimes (randomly) can recognize shapes from a set of random numbers. ||Organic shapes are so difficult that they are adapted towards random numbers||. superWow! // Per
  • Math Idea – Looking at Schrödingers equation. Every topic in math should be able to exist in a differential equation. Geometric diff. eq. Statistic diff. eq. Algebraic diff. eq. Programatic diff. eq. ?Analytic diff. eq.
  • Remember Einstein: He made classification problems. Cold fusion could be a property light emission problem. Put a different atom layer between two equal atom layers. That’s enough for a classification problem//P
  • Machine Learning Physics – Guess – Continuity Fusion? – I assume there exist a property effect when you got three layers with another atom type in between. Apply energy and does not want to interfere with the light emission. Otherwise a classification problem. -> Property fusion.
  • Machine Learning Physics Guess – With Quantum Loss functions maybe there are also Quantum Fusion. Since there is so much energy. A smart way would be to extract energy on a small scale. Quantum Fusion could be similar to glow plugs I guess.
  • Purpose of Materials give Innovation Clues – Guess – I wonder. If you know what the planet and ocean uses the material for. Could this information provide clues where we could use this material. By viewing the materials or atoms contact history. (in water?) // Per
  • Machine Learning Physics – Materials are models() and as such they could store knowledge in the form of loss functions. An advanced universe will have figured out how to store a lot of loss functions without losing information. Oil, water, wood, rocks, air.. all carry information
  • Simulating 77 with a list of target data points and a model in between. Difference between absolute error and squared. Both are ?not suitable for diff eq. Maybe a diff eq is a perticular loss function diffloss.

Archives

  • December 2019 (2)
  • November 2019 (46)
  • October 2019 (22)
  • September 2019 (18)
  • August 2019 (65)
  • July 2019 (45)
  • June 2019 (63)
  • May 2019 (92)
  • April 2019 (44)
  • March 2019 (43)
  • February 2019 (43)
  • January 2019 (78)
  • December 2018 (58)
  • November 2018 (57)
  • October 2018 (92)
  • September 2018 (55)
  • August 2018 (56)
  • July 2018 (42)
  • June 2018 (63)
  • May 2018 (35)
  • April 2018 (65)
  • March 2018 (69)
  • February 2018 (38)
  • January 2018 (34)
  • December 2017 (35)
  • November 2017 (43)
  • October 2017 (37)
  • September 2017 (23)
  • August 2017 (29)
  • July 2017 (31)
  • June 2017 (54)
  • May 2017 (45)
  • April 2017 (16)
  • March 2017 (25)
  • February 2017 (41)
  • January 2017 (38)
  • December 2016 (30)
  • November 2016 (28)
  • October 2016 (16)
  • September 2016 (33)
  • August 2016 (18)
  • July 2016 (25)
  • June 2016 (15)
  • May 2016 (34)
  • April 2016 (25)
  • March 2016 (16)
  • February 2016 (16)
  • January 2016 (25)
  • December 2015 (20)
  • November 2015 (12)
  • October 2015 (11)
  • September 2015 (5)
  • August 2015 (11)
  • July 2015 (11)
  • June 2015 (12)
  • May 2015 (12)
  • April 2015 (8)
  • March 2015 (6)
  • February 2015 (4)
  • January 2015 (6)
  • December 2014 (8)
  • November 2014 (4)
  • October 2014 (4)
  • September 2014 (5)
  • August 2014 (3)
  • July 2014 (2)
  • June 2014 (5)
  • May 2014 (2)
  • April 2014 (2)
  • March 2014 (1)
  • February 2014 (5)
  • January 2014 (1)
  • December 2013 (2)
  • October 2013 (1)
  • March 2013 (1)
  • October 2011 (1)
Rss Feed Link

site owner: Per Lindholm (peroglyfer@gmail.com)

About

Play Soundbite freewill

Proudly powered by WordPress
We use cookies to ensure that we give you the best experience on our website. If you continue to use this site we will assume that you are happy with it.Ok