AGI

search for “artificial general intelligence” in github and hit many results.

some article says deep neural networks are t2 level, human are t3 level (connected dnns) and to train a t3 level intel you need t4 level intel (like the earth, or a group of t3 intels, you can’t be 1-to-1 now, you must be forgiving). this proves my provision of putting agi into gui or human interfaces.

to train a t2 level intel, you need t3 intel, such as yourself or your users.

you must setup this playground then you begin to learn stuff and libraries of AGI. but once you have one, don’t move around, stick to it! you need time to develop the general adaptor and make clear and achievable goals!

terminal, GUI, program, API, network

please check my previous efforts on building AGI, namely: AGI, lazero, metalazero

nsjail with docker

firejail

said by HTM, AGI knows what it did to the world (self-awareness), also signals from sensors.

google research

gwern wrote a fiction. he thinks agi starts from automl-zero which is similar to lazero and metalazero by name and perspective.

by design lazero can be deeply aligned, inspecting and studying user’s actions. it also has its own exploration space. however, these expectations can never be fully satisfied at the same time. if you want more power, you have to let go.

lucidrains repositories

this one got lots of state-of-the-art implementations for close-sourced papers and also repos for AGI. stunning.

AGI related

JEPA-pytorch (WIP) yann lecun’s version how agi will be built

PaLM scaling language model with pathways

side projects

make a video text to video generation

nuwa text to video generation

opencog

moses (supervised) for evolutionary program synthesis

repos on github

he4o

aixijs general reinforcement learning in browser repo

opennars

brain simulator 2 on windows platform

materials and links

DQfD: Learning from Demonstrations for Real World Reinforcement Learning (paper)

mit class on AGI

jiaxiaogang’s god-knows-what theory and training logs

awesome deep reinforcement learning (deep-rl)

awesome agicocosci exhausitive list of papers and repos for cognitive science and AGI

introduction and links on AGI

review previous notes and fill blanks, list blanks below, better with direct link to it

• tag all notes, especially mark out those stub, incomplete ones

• review and complete bilibili courses, reorder, rename and split them if necessary

• 传播学导论把笔记做完

review your history

• visit all previously visited links and store briefs generated by readbility.js and elinks

finance and quantatitive trading

• design a basic algorithm and complete regression test on joinquant

• design and complete one regression test offline

artificial general intelligence

• design a program automatically execute commands in shell

• design a program automatically click everywhere in GUI

study nars

• run python version of nars

study opencog

study he4o