ALIEN CODES AND THEIR AUTOMATED AND HUMAN EXPLANATIONS
Slides: https://josefurban.eu/slides/scml26oeis.pdf . Abstract: We have automatically discovered symbolic explanations for over one-third of the sequences in the Online Encyclopedia of Integer Sequences (OEIS). The talk will describe the neuro-symbolic system consisting of a positive feedback loop that starts from zero knowledge and iterates between guessing the explanations, their verification, and training of the guessing methods. Then I will describe several additions and experiments that led to the current set of solutions found in hundreds of iterations of the feedback loop. I will show some of the solutions discovered. This includes over 80 programs for primes developed as the system self-evolves. I will also discuss a related experiment in automatically proving equivalences of the discovered programs using the SMT solver Z3. Because induction is often needed in such proofs, this leads to another self-learning neuro-symbolic system that repeatedly tries to guess the right instances of induction for Z3. Finally, I will also show some of the recent human explanations of the programs, mostly done by Tom Hales.

BYD: The Biggest Fraud in the Auto Industry

ISCA 2026 Tutorial 01: Introduction to Cryptolets

There's Something Weird Going On in M83

There’s a Problem with Quantum Mechanics – Quantum Reality (1/3) with Jim Al-Khalili

RL for Agents Workshop - Deep Dive on Training Agents with RL and Open Source

Training Sand to Think: Artificial General Intelligence & Future of Physics

Physics-Informed Machine Learning – Lecture 1 | Why Physics + AI?

The Langlands Program - Numberphile

Yann LeCun: World Models: Enabling the next AI revolution

The Hardest Questions in Physics | World Science Festival

Ben Wikler - “This Is The Plan” & Taking On Elon Musk In Wisconsin | The Daily Show

William Dunham, A tribute to Euler

But what is quantum computing? (Grover's Algorithm)

Full AI Prompting Course with Andrew Ng

Complete Terraform Course - From BEGINNER to PRO! (Learn Infrastructure as Code)

How AI Cracked the Protein Folding Code and Won a Nobel Prize

Is Iran Becoming Another 'Forever War'? (w/ John Mearsheimer) | The Chris Hedges Report

From Child Prodigy to Winning Fields Medal, Nobel of Math
![Yann LeCun's $1B Bet Against LLMs [Part 1]](https://i.ytimg.com/vi/kYkIdXwW2AE/hqdefault.jpg?sqp=-oaymwEjCNACELwBSFryq4qpAxUIARUAAAAAGAElAADIQj0AgKJDeAE=&rs=AOn4CLDbV4izF3i-wxevCVIn7FJjoy1vlA)
Yann LeCun's $1B Bet Against LLMs [Part 1]

