morgangiraud / fundamental-bias-collection Goto Github PK
View Code? Open in Web Editor NEWThis is a collection of potential and proven fundamental bias existing in our universe
This is a collection of potential and proven fundamental bias existing in our universe
I've encountered this idea reading this blog post from Max Welling: https://staff.fnwi.uva.nl/m.welling/wp-content/uploads/Model-versus-Data-AI-1.pdf
To give a little bit more context, it has to be read in response to the blog post of Richard Sutton here: http://www.incompleteideas.net/IncIdeas/BitterLesson.html
I'm more on the side of Richard Sutton on this debate, whatever the world is, it looks like compute power and engineering will help us solve everything.
The last example that computes beats us (outside of things like AlphaGo) is this paper. CNNs happens to be challenged by random neural networks.
Need to watch:
https://redwood.berkeley.edu/seminars/john-baez-march-2019/
https://en.wikipedia.org/wiki/Second_law_of_thermodynamics
Summary:
Given an isolated system (a system which can't have any interaction with other systems), the entropy of that can only increase.
Entropy :
To definite it clearly, we must first define a self-sufficient set of quantities that define our system at a macro-level.
Now the entropy of a system is proportional to the natural logarithm of a number of microstates which would result in the same values of the measurement of the macro-level quantity.
If you give a uniform probability over all the possible microstates. This is equivalent to saying that the entropy will end up being the most probable macrostate.
Implication: Unless you give energy to a system, it will never naturally evolved into a less probable state than the most probable one which is also the less ordered one :
One thing to note is that solid is more organised and so less probable than liquid which is itself more organised and so less probable than gas.
Also, the second law says "can only increase". This is because, if the isolated system is in equilibrium, nothing happens but if it's not, the entropy will increase.
Create an isolated system with an ice cube and a lot of heat, you'll end up with a hot gas
The main paper which explores auto-curriculum: https://arxiv.org/pdf/1903.00742.pdf
The paper: ftp://ftp.idsia.ch/pub/juergen/everything.pdf
I'll let Juergen Schmidhuber explain this one:
The universe is deterministic, and the most efficient program that computes its entire history is short and fast, which means there is little room for true randomness, which is very expensive to compute. What looks random must be pseudorandom, like the decimal expansion of Pi, which is computable by a short program. Many physicists disagree, but Einstein was right: no dice.
There is no physical evidence to the contrary (see Don't forget randomness is still just a hypothesis).
For example, Bell’s theorem does not contradict this. And any efficient search in program space for the solution to a sufficiently complex problem will create many deterministic universes like ours as a by-product. Think about this.
See Computable Universes & Algorithmic Theory of Everything: The Computational Multiverse and the talk here http://www.kurzweilai.net/in-the-beginning-was-the-code
A declarative, efficient, and flexible JavaScript library for building user interfaces.
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
An Open Source Machine Learning Framework for Everyone
The Web framework for perfectionists with deadlines.
A PHP framework for web artisans
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
Some thing interesting about web. New door for the world.
A server is a program made to process requests and deliver data to clients.
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
Some thing interesting about visualization, use data art
Some thing interesting about game, make everyone happy.
We are working to build community through open source technology. NB: members must have two-factor auth.
Open source projects and samples from Microsoft.
Google ❤️ Open Source for everyone.
Alibaba Open Source for everyone
Data-Driven Documents codes.
China tencent open source team.