Learning
Learning
The Great Course - Has Maths, Science, Electronics, etc...
Unified Mindfulness CORE Training Program
Right Click -> Links -> MooCs
[Filter\ NASA Publications]
How to fix fancy electronics in newer tools
Perovskite solar cells made simply
Open2Study - Concepts in Game Development
Misc
https://en.wikipedia.org/wiki/Transition_town
https://www.engineeringforchange.org/
appropriate technology library
Rendering Course - By the two minute papers guy. Should cover global illumination...
MOOC hubs
Robotics
Geometry
Introduction to Geometry - SchoolYourself
Microprocessor Fabbing
Maskless UV lithography for microfabrication, paperdump
E-Beam Lithography, Part 1 - Talks about the specific resist used and other details. Small desktop lithographic machine.
Nanometer Pattern Generation System. SPIE Handbook (Seems it might be a 'mod' for an electron beam). "Scan coils". "Beam blanker". 10min for 40x40micron. Scanning capacitance microscopy (SCM). Beam measured in picoAmps. 100k magnification. Uses a 'cup' to measure beam amps. A 'gold standard' is used to calibrate beam. Chip isn't going to be perfectly flat, look at the corners to know the height to focus, raise the stage.
Youtube Video? 2 - Unwatched
Photolithography Overview for MEMS - Unwatched
Intel Talk - Unwatched
DIY Semiconductor manufacturing In a box
Casting
‘quick view’ comparisons of casting materials - Polyester resin is cheap.
Make silicon mold out of clear silicon + cornstarch Auger Mixer for Oogoo
HydroCal (Is this gypsum?), whats the amount the powder makes?
3D Printer
Dollo github National Pipe Thread / British Standard Pipe
Electronics
3d Printing
Physics
Optics
https://en.wikipedia.org/wiki/Nonimaging_optics
Scala
Functional Programming Principles in Scala
Functional Program Design in Scala
Big Data Analysis with Scala and Spark
Statistical Learning
Finance
Complex Systems?
Math
Programming
Dave Akley - Robust First Computing
GameDev
Part I: Client Side of 64 Network DO’s and DON’Ts for Game Engine Developers
MachineLearning
- Deep Learning Patterns
- Deep Learning Gallery
- Awesome Deep Learning Papers
- NanoNets: How to use Deep Learning when you have Limited Data
- Deep Learning on medium.com
- NIPS2016 - Videos!
- ai-software-learns-to-make-ai-software
- KDD2016 talks
- Center for Brains, Minds and Machines (CBMM)
- Nando de Freitas
- OpenAI.com - GANs - Read This
- A Course in Machine Learning - Free Book
- NIPS2016 talks
- This has lots of good infos - Includes an AndrewNG talk and recommended papers.
- Reddit best papers
- Yes you should understand backprop - CS231N lecture on backprop (claims to focus on intuition)
- Wide Residential Networks Impl - 3rd party - keras impl
- Learning AI if you suck at math
- CNN Tricks
- RNN CS188
- Neural Networks and Deep Learning - Free Online Book
- An introduction to Generative Adversarial Networks (with code in TensorFlow)
Ideas
- Would it be possible to train a network on cut up sections based on how much they cause the neurons to spike? Don't train sections that have a big response already.
- Would it be possible to modify the inputs to 'censor' bits that cause over-fitting.
- Would it be possible to 'move' and object across the view overtime and learn that it's still the same object as a kind of data-augmentation?
- Could an auto-encoder be used to synthesise convolution filters for a pre-initialisation?
- Is it possible to learn a fitness function by starting in the goal state and trying to learn how to leave it.
- Then learn how to leave that state for a new one.
- Would be easier in a discrete, reversible, deterministic world.
- Would need to define how different 2 positions need to be to be considered different states.
- Is is possible to learn what a 'state' is?
- by taking 2 'non-goal' states and learning to move between them without triggering the goal and using a state halfway between as a new state?
- By taking a bunch of 'non-goal' states and and finding the maximum difference between them?
- Or using the distance between goal and non-goal?
- Or by using the distance between 2 non-goal states?
- Need to be able to reverse the 'move out of goal'.
- If the actions are reversible and deterministic then just undo them.
- Could relearn how to get from the state to the goal
- How to determine how far away a non-goal state is? How much time/how many actions it takes to get to the goal state from the non-goal state?
- What about a key/lock/door puzzle
- By default it wouldn't learn to put the key in the lock as the puzzle would either start in a solved state (or actor would be stuck behind the door).
- Could start the puzzle solved and make it become unsolved as the actor walks backwards, ie you pick up the key when you go though the unlocked door which becomes locked, then have to 'loose' the key it in the place where it's actually obtained. Is this just turning into as complex a problem as solving the puzzle in the first place?
Courses
- Deep learning at Oxford 2015 - Nando de Freitas (Bad audio quality)... (same guy did this talk)
- Foundations of Machine Learning Boot Camp + Deep Learning Specific stuff
- Neural networks class - Université de Sherbrooke - via Reddit "A complete and easy to follow course for understanding ANNs". Seems to have some math. Short <15min lectures.
- undergraduate machine learning at UBC 2012
- Machine Learning 2013
- videolectures.net
- Berkeley - CS 294: Deep Reinforcement Learning, Spring 2017
- Berkeley - CS 294-129 10/5/16
- Berkeley - CS 294-131: Special Topics in Deep Learning Fall, 2016
- Stanford CS231n Winter 2016: Convolutional Neural Networks for Visual Recognition. - Convnets in pratice 15min in talks about retraining just a littlebit..., at <36min talks about convolutions with 3x1 and 1x3.
- CS231N forums
- Coursera - Machine Learning - Andrew Ng
- Coursera - Neural Networks - Geoffrey Hinton
- Stanford - These tutorials seem cool
- Optimization Algorithms in ML - NIPS2010 - Video (2010)
- Kadenze: Creative Applications of Deep Learning with Tensorflow - Course seems nice.
- Kadenze: The Nature of Code
- Udacity - Machine Learning
- Udacity - Deep Learning
- TensorFlow - "Best Tensorflow+Deep Learning tutorials on YouTube"
- MIT - Intro to Deep Learning
- University of California, Berkeley - CS118: Artificial Intelligence
- MIT 6.S094 - RNN and control topics.
- Oxford - Deep Learning NLP
- Connections between physics and deep learning - Center for Brains, Minds and Machines (CBMM)
- deeplearningweekly.com - List of opensource curriculum. Lots of courses.
- deeplearning.net Datasets
- OpenML - Has heaps of datasets.
- HyperNEAT - HyperNEAT is a well known Neuro-Evolution algorithm
- Learning Genetic Algorithms - Guy with candy.
How to Install OpenAI's Universe and Make a Game Bot using reinforcement learning.
fast.ai
- Lesson 3: At <1:18:00 shows how to manipulate and fine tune a model. Says always use batch normalisation (1:40:00). Mentions the BatchNormalization layer that does batchnorm for you. Shows making a model from scratch and Ensembeling them (1:57:30).
- Finished Lesson 4 - (Geoffrey_Hinton says max pooling is bad. Talks about capsule architecture) - Use ADAM, look into the Jeremy Howard's modified ADAM. optimizer=Adam() (model.compile), Previously he talked about always using RMSprop...
- Lesson 6: Says loss="sparse_categorical_crossentropy" (model.compile) allows you to avoid one hot encoding (1:26).
- Lesson 7: 30min in he talks about some kdd best paper competition. Finding bounding boxes at 49min in.
7 Steps to Mastering Machine Learning With Python - "From classification, we look at continuous numeric prediction"
Carnegie Mellon University Course
Stanford Machine Learning Unofficial Notes
The 10 Algorithms Machine Learning Engineers Need to Know
Machine learning for algorithmic trading w/ Bert Mouler
neural aesthetic @ schoolofma :: 10 convnet applications
Royal Society on Machine Learning
Evolving AI Lab EvolvingAI.org
DeepLearning.TV - Watched as on 2016 Aug 13
Tensorflow and deep learning, without a PhD, Martin Gorner, Google
TensorFlow
Deep Learning With Python & Tensorflow - PyConSG 2016
Intro to ML and TensorFlow Tutorial
The Ultimate List of TensorFlow Resources: Books, Tutorials, Libraries and More
Tensorflow: How to restore a previously saved model
PyTorch
Practical PyTorch: Classifying Names with a Character-Level RNN
Misc
- Modulus layer? +1.0 -> -1.0 (modulus is apparently expensive...)
- Prune out neurons that don't fire for many images as a way of regularization?
Color Representation Ideas
- See what effect reducing the bits of colour has on accuracy... GreyScale vs R8G8B8 vs R8G4B4, etc...
- Dumb & basic - Red*255*255+green*255+blue. Should reduce the number of channels but still have precision bits left over. Maybe it needs to be offset to the centre though to help future layers...
- Would it be possible for a Float32 to be used to encode 8 bits of the next color?
- AllRGB - Shows images with one of every colour in it...
- The way the human eye splits things up? That's like 6 separate images...
- More natural mapping...
- Maybe try and keep similar colours together distance wise? Hard to do with a single dimension float.
- Hilbert Curve
- Periodic algorithms? - Sin,Cos,etc..., Repeating patterns, Partial gradient+periodic. Isn't RGB basically just periodic anyway?
- Float32 using 4d color space (but then some colours would be duplicated...]
- 2xFloats merge red&blue?, green&red? (how does the eye do that?)
- Separate out brightness?
- Try an optimisation function similar to an embedding?
- Maybe try and make similar colours further apart to exaggerate the subtle differences?
- Negatives could cause the 'near' dimensional to change... Like having an extra dimension. Would duplicate colours again.
- What about that Kaggle competition with the 16 band sate-lights?...
- Maybe this is all useless. The convolution layer should just learn what it needs anyway... It's all probability based so it shouldn't need to be too close. This could all hurt it.
Distributed
How To Create A Fairly Decentralized Commenting System
Computer Science
- MIT: Introduction to Algorithms (Fall 2011) - Part 1/3. Θ
- Up to Lec 5, 22min (also watched the last 2)
- MIT: Design and Analysis of Algorithms (Spring 2015) - Part 2/3
- MIT: Advanced Algorithms (Fall 2008) - Part 3/3
- MIT: Computational Geometry: (Spring 2003) - No videos :(
- Computational Geometry Lecture 1: Review of linear algebra
- MIT: Distributed Algorithms (Fall 2009)
- MIT: Structure and Interpretation of Computer Programs (Spring 2005)
- Computation Structures 3: Computer Organization - Starts - Mar 1, 2017
Finance
Scrape from PDF?
Marge CUA emails with paypal ones (but not all will be paypal payments and the CUA emails don't give details, probably useless).
Implement Vault (passwd?)
Reactive
https://github.com/evancz/elm-architecture-tutorial/
http://elm-lang.org/ via https://news.ycombinator.com/item?id=10746533
Talks
21st Century Software Testing, David MacIver, Hypothesis
Clean Code
Open Hardware Summit 2014 - Rome
Fab Academy 2016 Recitations - Nadya Peek
30C3: Making machines that make (EN)
Programming a new reality | Neil Gershenfeld | TEDxCERN
Towards General Artificial Intelligence (Google rat level AI)
Decentralised Web 1:15h day 2
This guys uploads! Also this guy
The introduction to Reactive Programming you've been missing
Railway Orientated Programming
Domain Driven Design with the F# Type System
Design Patterns in the Light of Lambda Expressions
Go: Object Oriented and Concurrent (just not the usual way)
Future Programming Workshop - SPLASH