Dirk Roeckmann

Computer Scientist, SAP Expert, IT Manager, Consultant, Independent AI Researcher, AI Engineer 👉 AI = Deep Learning + Causal Inference + Symbol Manipulation

[Dirk Roeckmann @fivetroop]

AI = Deep Learning + Causal Inference + Symbol Manipulation A I = ∇ w + P ( Y = y | d o ( X = x ) ) + λ
#AI #GenAI #CausalAI #KnowledgeGraphs #NeuroSymbolics

[Dirk Roeckmann @fivetroop]

@AiwithDharmik 93

[Pedro Domingos @pmddomingos]

Reminder: all that gradient descent does is store memories in weight vectors by superposition.
https://t.co/irTkCbj2XC

[Gary Marcus @GaryMarcus]

💯💯💯: “If AI is going to produce trustworthy and genuinely novel insights in science, medicine and beyond, we need systems whose conclusions arise from an auditable process of evidence, inference and belief revision.”

It is madness to believe otherwise.

[Pedro Domingos @pmddomingos]

In the Middle Ages only the priesthood could access the Word of God because only they could read Latin. Now only the AI priesthood can access the Mind of the Superintelligence because only they can interpret its weights. But the printing press is coming:
https://t.co/ZbJDGY7GOL

[Dirk Roeckmann @fivetroop]

Instead of #Python, #JS etc, think #Gleam #OCaml etc and even #Lean4 without using proving capabilities (think ‘partial’) for #agenticcoding loops. Discrepancies in training data size can be alleviated by creating synthetic code. These langs give top feedback at compile time. #AI labs listen
3/3

[Dirk Roeckmann @fivetroop]

Since #LLM as well as harnesses are pretty good in the #coding domain nowadays, we have to focus more on the programming languages themselves in terms of suitability for #agentic #recursiveselfimprovement loops. Current favorites are mainly ‘favorites’ because of the size of the training dataset
1/3

[Dirk Roeckmann @fivetroop]

Dynamic typing, mutability, object oriented programming are all bad choices for agentic feedback loops. #Functionalprogramming languages with #algebraicdatatypes, #immutability are more viable because they avoid classes of errors and provide detailed compile-time feedback.
2/3

[François Chollet @fchollet]

The narrative around why building AGI was a good idea has always been about invention -- "it will cure cancer", "it will solve fusion energy", etc.

So I would not "declare AGI" until we have AI that is, at last, capable of invention -- conceptual breakthroughs, novel insights, new real-world technology, etc.

[Dirk Roeckmann @fivetroop]

@fchollet @analisereal What about #SeedIQ running on a M1 laptop scoring 100%!?

[Dirk Roeckmann @fivetroop]

@VisimoD @fchollet @GaryMarcus The whole harness is purely symbolic

[Dirk Roeckmann @fivetroop]

@AnatoliKopadze Yea right. We started with prompts. Now we write loops. Then we write algorithms. Then we’re back at programming. Wait a minute … 😂 #AI #SAP #ABAP

[Dirk Roeckmann @fivetroop]

Food for thought: #LLM are black boxes, so are humans, but we deal with them seeing them as non-/cooperative player #markovblankets and using our perception/action loop to probe them. Let’s do that with #GenAI too. There’s your path to explainability.

[Dirk Roeckmann @fivetroop]

@fchollet So we stochastically evaluate the quality of perfectly deterministic algorithms? What’s the motive?

[Dirk Roeckmann @fivetroop]

@fchollet @fchollet - why is Seed IQ, the only non-human entity with a 100% score never mentioned? @deniseholt_

[Dirk Roeckmann @fivetroop]

#SAP has to get their API act together. Prohibition never reduces but always increases the demand of what has been prohibited. This is ethically wrong, legally shaky, politically disastrous and economically ill-advised. What’s not to like? 🤦😂
2/2

[Dirk Roeckmann @fivetroop]

#SAP, with the latest update of its API policy threw customers and the community into turmoil by trying to stifle ways how users can access THEIR code and data and expose it to non-SAP tools and platforms of THEIR choice. Clarification is needed ASAP including a white/blacklist of #APIs.
1/2

[Dirk Roeckmann @fivetroop]

@chrishayuk Watched your #larql video and it got my gears turning. Does this mean it could be possible to ‘latch on’ attention to an #ontology?

[Dirk Roeckmann @fivetroop]

We have to support ‘biodiversity’ of programming languages.We can’t let #agentic #coding solidify 2 or 3 top languages just because they make up the majority of training data and neglect all other promising ones.This stifles progress of the very substrate of executable formal statements. #SAP #ABAP

[François Chollet @fchollet]

Simply retrieving a reasoning trace looks a lot like human reasoning, until it's time to navigate uncharted territory. If you memorized all reasoning traces of humans from 10,000 BC, you could automate their lives but you could not invent modern civilization.

[François Chollet @fchollet]

The reason symmetry is so important in physics is because symmetry is a highly effective compression operator. If a system is invariant under some symmetry, you only need to explain one axis of it. Scientific models represent the systematic exploitation of the universe's internal redundancies through symbolic logic.