I made my first post to the #AI section of the arXiv this week! You can find the preprint "Discrete neural nets and polymorphic learning" at https://arxiv.org/abs/2308.00677.
In this paper a learning algorithm based on polymorphisms of finite structures is described. This provides a systematic way to choose activation functions for neural nets whose neurons can only act on a fixed finite set of values. These polymorphisms preserve any specified constraints imposed by the learning task in question.
This paper is the result of a 2021 REU at the University of #Rochester. I am working with a great group of students on a follow-up project right now, so videos of talks and a sequel preprint should be out soon!
#NeuralNets #MachineLearning #math #combinatorics #UniversalAlgebra #ComputerScience #teaching
#teaching #computerscience #universalalgebra #combinatorics #math #machinelearning #neuralnets #rochester #ai
The Godfather in Conversation: Why Geoffrey Hinton is worried about the future of AI
https://www.youtube.com/watch?v=-9cW4Gcn5WY
#Hinton #mustsee #AI #KI #ChatGpt #neuralnets #Backpropagation
#hinton #mustsee #ai #ki #chatgpt #neuralnets #Backpropagation
With all the talk about AI right now. If you're interested to learn about the basics of Neural Networks, check out one of my older talks at Node JS Interactive.
(Disclaimer: This talk is over 4 years old, and I was severely jet-lagged ๐ฌ)
#ai #neuralnets #neuralnetworks
Long read investigating what is going on behind the scenes with #ChatGPT #ArtificialIntelligence #NeuralNets #Modeling https://writings.stephenwolfram.com/2023/02/what-is-chatgpt-doing-and-why-does-it-work/
#modeling #neuralnets #artificialintelligence #chatgpt
If you're a techy like me but not specifically in the language / neural net / AI world, then you're probably a bit bemused by ChatGPT.
Set aside an hour or so (it's a long read!), and read this excellent writeup of how it fundamentally works, from Stephen Wolfram
https://writings.stephenwolfram.com/2023/02/what-is-chatgpt-doing-and-why-does-it-work/
It's a big system, but some of the guiding principles within it are fairly simple.
#chatgpt #ai #neuralnets #stephenwolfram
I got the basic thing working. I also have a script to convert my icon images into arrays of floats, so almost ready to train it for real.
First attempt will be just 2 hidden layers of 64 neurons. I expect this to be good enough. But can be fun to later follow the other tutorials and make it more fancy.
More long term I want the app to be able to interpret images of wireframes. That requires a more advanced network..
People like surveys :) So here is another one:
You're learning about #computervision foundations (no #NeuralNets :woohoo: ) and you want to implement some algorithms by hand.
What language do you use?
Who was Marvin Minsky? Article written by #chatgpt #AI #DeepLearning #NeuralNets https://open.substack.com/pub/vzocca/p/who-was-marvin-minsky-by-chatgpt?r=1nqhf3&utm_medium=ios&utm_campaign=post
#chatgpt #ai #deeplearning #neuralnets
All this hype about AI art has made me want my own AI!! #gamedev #neuralnets #coding #ai
#ai #coding #neuralnets #gamedev
Data science & Groovy using #ApacheBeam #ApacheCamel #ApacheCommons @ApacheGroovy #ApacheIgnite #ApacheMXNet #ApacheOpennlp #ApacheSpark #ApacheWayang #Datumbox #deepjavalibrary #DeepNetts #EclipseDL4J #graalvm #gradle #stanfordnlp #TensorFlow #Tribuo #smile #tablesaw #opencsv
https://www.javaadvent.com/2022/12/groovy-and-data-science.html
Covers data manipulation, regression, clustering, classification, natural language processing & object detection #jsr381 #visrec #ai #ml #groovylang #neuralnets #deeplearning
#apachebeam #apachecamel #apachecommons #apacheignite #apachemxnet #apacheopennlp #apachespark #apachewayang #datumbox #deepjavalibrary #deepnetts #eclipsedl4j #graalvm #gradle #stanfordnlp #tensorflow #tribuo #smile #tablesaw #opencsv #jsr381 #visrec #ai #ml #groovylang #neuralnets #deeplearning
Data science & Groovy using #ApacheBeam #ApacheCamel #ApacheCommons @ApacheGroovy #ApacheIgnite #ApacheMXNet #ApacheOpennlp #ApacheSpark #ApacheWayang #Datumbox #deepjavalibrary #DeepNetts #EclipseDL4J #graalvm #gradle #stanfordnlp #TensorFlow #Tribuo #smile #tablesaw #opencsv
https://www.javaadvent.com/2022/12/groovy-and-data-science.html
Covers data manipulation, regression, clustering, classification, natural language processing & object detection #jsr381 #visrec #ai #ml #groovylang #neuralnets #deeplearning
#apachebeam #apachecamel #apachecommons #apacheignite #apachemxnet #apacheopennlp #apachespark #apachewayang #datumbox #deepjavalibrary #deepnetts #eclipsedl4j #graalvm #gradle #stanfordnlp #tensorflow #tribuo #smile #tablesaw #opencsv #jsr381 #visrec #ai #ml #groovylang #neuralnets #deeplearning
After the initial excitement has set by now, more and more people have raised the issue that the GPT-3 based chatbots are able to make statements that sound convincing, but may have no bases in reality.
This makes a lot of sense to me, as AI, in its core, is based on one thing: pattern recognition.
So really, all it does is cargo cult, not learn or necessarily understand.
#gpt3 #chatbot #ai #ml #neuralnets
Using #ChatGPT3 as the world's most inefficient #Linux machine ;)
More craziness here: https://arstechnica.com/information-technology/2022/12/openais-new-chatbot-can-hallucinate-a-linux-shell-or-calling-a-bbs/
#chatgpt3 #linux #ai #neuralnets #gpt3 #openai
Are the #neuralnets afficionados (re)discovering the most computationally expensive way to do #Bayesian #statistics again ?
(There are some other goodies about the Jensen Shannon divergence in Wikipedia article , one smartphone screen up but will not bother you with the details)
https://en.wikipedia.org/wiki/Generative_adversarial_network?wprov=sfla1
#generative #adversarial #AI
#neuralnets #bayesian #statistics #generative #adversarial #ai
Finally! I adapt the implementation code of a paper from my project and now on #Github and #PapersWithCode :partyparrot:
The article that I implemented is about #neuralArchitectureSearch #neuralnets #geneticAlgorithms
#github #paperswithcode #neuralarchitecturesearch #neuralnets #geneticalgorithms
@esteng some images for context.
- colored one: sketches to figure out how to do the one-hot-encoding for shapes of diff size.
- grid: gerated wireframes. Only couple of shapes, but lots of positions, sizes.
- the actual UX app: It has a lot more shapes, etc. Canvas is wider than single wireframe.
- graph: results training on 4x4. But larges sizes and more shapes quickly became much harder.
Manipulation of user #demographics on #socialplatforms by #neuralnets / #machinelearning /a.i. should be regulated and outlawed.
#demographics #socialPlatforms #neuralnets #machinelearning #twittermigration #facebook
Can anyone tell me the benefits to use #CartesianGeneticProgramming instead of #TreeBasedGeneticProgramming to #neuralnets architecture #search ?
#cartesiangeneticprogramming #treebasedgeneticprogramming #neuralnets #search
I started with #MachineLeaning then I realize that I need #math (I start studying it). Now Iโm learning the foundation of #AI like searc algorithms and different learning algorithms exluding #neuralnets โฆ I feel I started in the end and Iโm going to the beginning ๐
#MachineLeaning #math #ai #neuralnets