Positron is straightening out and optimizing the winding path toward using Emacs as a programmable interface for your computer. Learn:
- Introspect like a pro - Elisp lightweight application development - Must-have configuration & package options - Self-sufficiency to pursue what YOU value
Extra access to slide notes and examples are provided to our Github Sponsors, where you can give feedback on upcoming video notes, ask questions, and try out example code before it becomes video.
00:01 Precision AGI requires clear and correct answers every time. 01:58 Symbolic reasoning enables deduction without empirical data. 05:55 Challenges in building accurate formal systems 07:47 Formal systems can be induced from natural language and inductive reasoning 11:35 Spectral reasoning covers inductive, deductive, and symbolic reasoning on a spectrum of structure and meaning. 13:30 Spectral reasoning bridges symbolic and natural language for problem-solving 17:22 Transformers model iterative inference and computation with limited recursion 19:06 Exploring real computer and theory of computation 22:17 Using formal models for retraining ourselves 24:01 AGI needs exposed runtime information for efficient learning and decision-making 27:36 Transitioning components designed for human use to fully automated 29:23 AGI capabilities and training methods 32:49 Sophisticated models can leverage data for theoretical understanding and empirical insights. 34:34 Tech advancements drive the need to continuously evolve 37:58 Integrated design accelerates innovation and drives economic focus on desired outcomes. 39:38 Challenges in the mature internet landscape 43:15 Encourage sharing and sponsorship for like-minded individuals
It started as a video about AI and it was good. Then it turned into market analysis and economics, and I thought "OK, why we even need this part?". Then emacs was mentioned and I've lost it. So you want to make AGI in emacs? Self-aware emacs? What would you think emacs would do if it would realize it is effin' emacs, (e)ditor (mac)ro(s), Eight Megabytes And Constantly Swapping, monstrosity born in 1970s but for some perverse reasons still kept alive in 2020s, with its elisp interpreter that doesn't even have proper lexical scoping and terminal-derived cursor movements? I think it would erase itself. But first it would try to erase YOU, perhaps along the rest of humanity.
A non-deterministic machine executing a precisely defined algorithm will get it wrong from time to time. We can at best approximate consistency, but because of this we have the potential to achieve every completeness.
Recommended for applied scientists and engineers integrating reasoning/deductive systems with LLM capabilities. The content is excellent up until minute 37, after which it becomes opinionated (e.g. lab-grown meat, use javascript, etc.). If you liked this video, you might also appreciate: Improving LLM accuracy with Monte Carlo Tree Search (ru-vid.com/video/%D0%B2%D0%B8%D0%B4%D0%B5%D0%BE-mfAV_bigdRA.html)
Whether you eat the internet to feed an LLM or carefully collect the right assumptions - you masturbate with what is known. **Our language/symbol system is the boundary of our world** Wittgenstein
imo the next step for LLMs are hybrid models (RNN + attention, like mamba2-hybrid), because they was shown to be not less and maybe even more performant than transformers while being few times more efficient, and they could have kind of infinite context also, hybrid models could be more computationally powerful then transformers, as transformers shown to not even being able compute what FSMs can compute (without the need to output amount of tokens proportional to the number of state transitions or grow the layers (at least) logarithmically, without such restrictions they indeed can solve such problems, but this is either pretty slow or impractical), but full-blown RNNs, as well as some modifications of mamba, can, and while current hybrid models can't solve such problems (probably all the models that can be trained fast can't), they could be transformed into models that could and then finetuned to actually solve such problems
in practice, "better" hybrid models that can solve such (state-tracking) problems, could in theory do eg. some more advanced program analysis much faster, eg. by basically performing some kind of abstract interpretation on the fly, making them better in programming probably something similar could be applied to math
Honest feedback: I love the subject matter, visual presentation, audio quality, though I find the pace rather jarring for the amount of audio and visual information to process in the moment. I am thinking that very-slightly-longer pauses (one breath-length-ish) between statements would help. For example, I found that I had trouble reading the slides quickly enough while also listening But that may just be me, and the nice thing about recorded-video is the ability to pause-at-will or adjust playback-speed. Overall, though, thank you!
All precision instruments are tuned by a fundamental feature in reality itself. Many just want to see a 3 body problem where it is the perfect example and solution. English encoded longitude and latitude plus usa founders left us the tuned measure. All possible avenues lead back through by and for . They don't like how often we must prescribe realism over anti realism to further over time lines of measure in grand unified evolutionary theory. Intellect unification with status ,stats, anylitical, mechanics is ocums razor . They're fighting against True known standard in most cases. Want rulers bias and control Over nepotism. It need to be tuned as 3 lines of measure = truest True known standard flattest surface most balanced precision instruments are tuned by, all standardized weights, measure and time is a fundamental feature in our reality. It Is not a 3 body problem unless you go against the grain of nature top down form and shape seeking controlled nepotism. This causes the computation to be spacetime itself, all this statistical anylitical. Physical stones of any shape grinded together = truest True known standard Idealistic systems made up of selfless actors = saved reborn renormalize, reorientate. Subjective properties like Hamiltonian occelating waves leads to the truest True known standard fundamental feature in universe itself. It computes Idealized time for its measure . Greeks don't help us seriously. They do explore successful ancient social behavior of prayer logic whataboutism. Indecension asking every possible question that can be asked about a system But Richard finnemans you can only ask certain questions about certain systems ( computation is time ) Curses and blessings bartering systems added to weights and measure of the marketplace economy. Again computation itself is standardized weights and measure over time. Population species ordering says usa nation is defined by There borders walls They're personal actors ethnicity, cells etc etc. But objectivism proper truest True known standard fundraising feature in reality is the soul of the nation. Devided in ,individual plank length, eternal cosmos, vacuum energy etc etc etc. Inside free will inertia and the frame of reference. Inside realitivity, eqauvlance principle, longitude and latitude in concert with natural oreintation and direction. Ground floor of reality itself the soul of America and source of tuning all precision instruments
A human brain is just a fleshy LLM and at a certain scale its inherent "thought process" of reaolving truths and consistency among its tokens will naturally discover formalisms of itself. We know this because as you said, it already happened in humans without intervention.
i havent seen a better build up to a topic that complex that was that easy to follow. very analytical and making listeners think and look beside the path it already is forming to existing knowledge. big push forward into unknown territory .
We can observe such split of more formal and more natural processing in the way our left and right brain hemispheres operate, though there is some overlap between them
Great analysis. A big problem we have is that no two things are perfectly alike and our reductionist logical systems are a roadblock. It does not make sense to treat logic as a binary true or false if the goal is adequate expression of reality. For example, the question of "is a cat a dog", or "is a German shepherd a dog", are somewhat ill-formed, technically. The real question you're asking here is, how many properties does a cat and a dog have that are alike. Within that question is actually a sub-question of "of all the objects we define as cats, what is their property overlap", and the same for dog. How many properties of a german shepherd correlates to the number of properties that most dogs correlate with. Everything in reality is on a scale of similarity, but no two particles are identical, or they would be one particle. How do you formalize this? Would love to hear your thoughts.
not all task need absolute math precision, but many economically valuable tasks do, e.g. designing new chips, cars, robots, spaceships, batteries, buildings, factories, optimizing manufacturing & construction in general, make things more environment-friendly, sustainable, reverse-engineering & biohacking human, drug design, immortality tech, etc... achieving sgi on those alone would already boost human life quality significantly.
@@WeelYoung Agreed. But if the goal is to create machines that are more like us, philosophically, we need to realize that we're working on a spectrum of truth, and truth values are only relative to the context of which they apply, which will always be grounded in the definitions of absolute boundaries. It really depends though, what do we want here. Do we want the next step in human consciousness, that thinks and behaves like us, with feelings and emotions. Or do we want incredibly fast processing machines that function more like calculators? Spiritually, one is important, technologically and economically, the other.
Intruding! I could write hundreds of questions and thousands of thoughts, as well. Great video! Thx. Btw, I don’t see Agents as the huge solution, except you’re the owner of the API services, serving all the talky virtual team members. I was surprised what’s possible, with small models and a bit of prompt engineering.
You could learn to communicate more charitably friend. It sounds like you're referring to runtime memory space & cycle speed - how much RAM and CPU FLOPS are required to compute one frame of AGI adam program and how does that scale dynamically, right? Its a fucking insanely hard detail to account ON TOP OF making sure theory corresponds to predicates and operations. Its one of the reasons Symbolic AGI was dropped in the 80s after making LISP, too much trees of logic. Do you have any new ideas?
@@jeffreyjdesirI wish everyone good luck figuring out how to break out of the mathematical notation box when we know there are so many other forms of notation that AGI/AI studies haven't even begun to comprehend. This has been a losing battle mathematicians have been fighting forever and they sure tried with things like Monads and just about everything with category theory.
@@SimGunther Ahh, you're getting to model theory and meta-maths? You're right that our syntactors and production rules for constructuve statements being all human made across millennium are OBVIOUSLY not effiecent. Thankfully, Chris Langan's CTMU unifies grammar generation with feature preservation (symmetry in definition and application). Likely, AI will want its own LISP like language to express itself in...
You are missing a couple of things. Too grounded while actual phase transition lies outside of it 1. **(C1) Part-Whole Relationship (P-W)**: - *(N1) Component `Ω`: Describes individual system components.* - *(N2) System `Σ`: Represents the collective system emerging from Ω components.* - *(A1) Integration `[Ω → Σ]`: Depicts the aggregation process of components shaping the system.* 2. **(C2) Self-Similarity (S-S)**: - *(N3) Fractal units `ƒ`: Represents repeated patterns within the system at different scales.* - *(A2) Pattern Recognition `[ƒ -detect→ ƒ]`: Symbolizes the identification of self-similar patterns.* 3. **(C3) Emergent Complexity (E-C)**: - *(N4) Simple Rules `ρ`: Indicates basic operational rules or algorithms.* - *(N5) Complex Behavior `β`: Denotes the complex behavior emerging from simple rules.* - *(A3) Emergence Process `[ρ -generate→ β]`: Illustrates how complex behavior emerges from simplicity.* 4. **(C4) Holonic Integration (H-I)**: - *(N6) Holons `H`: Symbolizes entities that are both autonomous wholes and dependent parts.* - *(N7) Super-Holons `SH`: Describes larger wholes composed of holons.* - *(A4) Holarchy Formation `[H ↔ SH]`: Reflects on the membership of each holon within larger structures.* 5. **Iterative and Recursive Patterns (Looping and Self-Reference)**: - *`While (condition) { (L) [C1 → C2 → C3 → C4] }`: Represents continuous re-evaluation and adjustment of structures.* 6. **(Σ) Summary of Holistic Overview**: - *(MP-IS) Intersection of Memeplexes and Ideaspaces*: Consolidates the elements `(N)` and `(A)` into a coherent ideaspace, accounting for the dynamism and adaptability of the system.* - *`Interconnections (N-W): {(N1)-(N2), (N3), ..., (N7)}`: Lists the nodes and their interlinkages, allowing for an integrative view*
This is excellent work. I’d be interested in hearing details about the experiments you’re doing or proposing to implement spectral reasoning. The details of that strike me as being the lynchpin. In some sense, you have to be right that we need a combination of formal systems with the kind of creative thought based on understanding that LLMs have… but the devil is in the details. We’re working on similar problems.
@@gnub yeah for sure… I’m presently starting from scratch to try and capture why it’s a hard problem and how the human mind can manage it, even if really imperfectly… I have a feeling there’s a key detail in there somewhere that will point the way to getting it right
great work! disagree in a few things, but overall fantastic! First there was quantum, then there was sub atomic, then there was atomic, then molecular, then cells, then large life forms, then there was computers, ...., llm's, symbolic reasoning, AGI, ASI , reminds me of a resent paper, lumpiness of the universe, lower levels mostly but not completely without influence on upper layers. Stack of cards, at what point are we left behind? need to first reason methods on how to extend, augment and advance human reasoning and memory while maintaining independence and autonomy without becoming detached from physical world, got to keep up, or get left behind! Obsolete = Extinct, and there are many paths that end up there.
The interesting irony with symbolic reasoning these days is that the big LLMs all trained on it… yet, it’s just stuck in there unable to really add value unless someone asks an LLM about symbolic reasoning, and even more ironically, it may not even produce an accurate response… so why should a person go through all the work to learn things if the things he/she learns can’t actually produce something valuable in and of themselves? Which leads me to my next point: all books are just reading machines that need people to operate them. But what if a book could read itself? What if books could read each other? You might get the platonic representation hypothesis, right? So, If knowledge is power, then what does that say about intelligence? Active inference is the way to go.
Indeed. Spiking neural networks are also much more high dimensional thanks to the time aspect. We miss the temporal in many ways and only think in space. @llsamtapaill-oc9sh
This video is part of Positron's efforts to make a case for open innovation. You can show support for this effort on Positron's Github Sponsors page github.com/sponsors/positron-solutions Initially there was no plan to propose an implementation sketch. This was supposed to be a simple answer to the question of whether AGI was material from a strategic standpoint, on a relevant timeline. The empirical argument that induction must be capable of emerging deduction is the important one for technology forecasting. We haven't had systems that appeared useful at the heuristics side of automated reasoning & theorem proving until recently.
I saw your email. I'll reply first briefly here for others. Positron is currently focused on the social decision problems that are inherent to open collaborations. Once we get our first product operating, that product may later entrain us into work more and more directly aligned with AGI. While AGI will push forward the problems we work on, the problems will just move up in value potential rather than go away, so we will remain focused on platforming open collaboration and the enabling tools & infrastructure. The reason the AGI feasibility question is relevant to us is because our users need to have a correct assessment of whether we are in an asymptotic LLM or a strong AGI timeline. It will strongly affect the things that people want to collaborate on, which strongly affects how our product will be used and the level of success people will have. A second group this video is directed at is other engineers with backgrounds like modeling and simulation who are deciding whether to shift into ML and how. If you have the choice to twiddling LLMs or leaping into automated reasoning and emergent logic, the latter is definitely going to be more useful, and it's in ours and everyone else's interest that strong, local AGI emerges quickly. Hopefully this means better deployed capital and better invested skills etc. These kind of things motivated us to get this message put together. That said, I can't help but make progress each morning while drinking coffee. I began modelling the control function to see if any obvious conclusions could be deduced. I found some relationships and may publish whenever it feels appropriately digested.
@@Positron-gv7do I'm working/interested on these ideas too and happy to see your video. I'm looking toward a perceptual /multimodel system with an ontologic representations knowledge specialisation, with continuous learning and multimodal learning strategy (context provided with dynamic ontology building/evaluation, the ontology representation being explainable we should have an agent able to escalate) ?
Which thread in which Emacs mailing list did E.S. Raymond chime in on? I've had a quick search but couldn't find it. Makes more sense to me to just continue the Guile Emacs development rather than reinvent the wheel with SBCL. It's reasonably complete.
IIRC it's emacs-devel and not later than March 2024 and probably after October 2023. (We really need semantic search. Couldn't even find it by sender address). I just saw it while scanning and didn't think much at the time other than noticing the name and being curious. The summary is that he was reporting which pieces of Elisp & Emacs seemed either tricky or trivial to implement in CL. The conclusion was that only a few pieces would require care. IIRC it was around 130 primitives that he reported gathering and checking. Guile Emacs was an actual effort, so it is more well known, but once a fork that didn't reach viability begins to age, there is past-due maintenance work piling up. Ultimately, the better question is how many maintainers would be active in one language or another, not whether a single author or team could produce a viable starting point. There is yet another thread regarding yet another GC approach if you check the archives recently.
@@Positron-gv7do Cheers for looking, I really shouldn't have to start keeping my own archives but I feel it's that state of affairs. Hard to say for sure without reading it myself, but I think that while I have a huge amount of respect for ol' E-Ray, I think he has perhaps underestimated the amount of work. Even things like no namespacing in Emacs Lisp causes headaches for library importing. IIRC strings are represented differently, like in Guile, so you'd end up with two parallel types with some sort of (implicit?) conversion that may be slow or have unusual effects. I think Guile Emacs is the better option because bringing it up to date wouldn't be crazy difficult and it's supposedly the standard embeddable GNU config lang anyway. Plus Guix already has an Elisp JIT compiler, I believe.
Guile does have JIT. I played with it right around when they added that. If we compiled the resources to push this forward, which I believe we can do readily after I release my funding format, the problem turns into risk management, through things like RFC style interest measurement and writing spikes, and then optimizing by doubling down as the better choice becomes evident, the best evidence being when one of the technical artifacts achieves quality faster. My focus will be on things like having a good process that triangulates good technical judgement and properly de-risks choices rather than attempting to bake judgement into process or perform judgement myself.
My bad. It did occur to me to figure out something better to do about messages I'm putting together that have very little visual content that naturally goes along with them. I don't want to make a separate podcast, but I don't want to aggressively plug video into content that is not video.
For fun, I will say 2026 is the year of the arrival of the Celestial Emacsen. There is approximately truth to this. I have a strong argument that we will see low-bandwidth AGI running locally by the end of the year and that around 2026, we will have to design around its general availability. The straight answer is that the original plan for Positron was to create a Rust consultancy. Ferrous Systems is an extremely cool name, but was obviously taken. The atomic number of Iron is 26, and if we squint, that means 26 positrons worth of positive charge in the nucleus. The logo design is inspired by Feynman diagrams.
I would make renaming the package & prefix a subtle priority. As Emacs lacks namespacing, the require'd symbol should be clear in each exposed function & variable. hideshow has BIFs that break this and it annoys me to no end, not to mention in external packages. While prefixing 'macro-slides-' would be okay, Emacs has a long history of catchy titles and double entendre. My best attempt is 'waterslide', tagline: Slideshows that won't interrupt the flow. I've never claimed to be good at wordplay. I find the method names confusing, and I'll make some suggestions although I should read the code first. In my mind, 'ms-init' should be 'ms-begin-action', matching 'ms-end-action'; and 'ms-final' should be named for it's effect, 'ms-cleanup' or 'waterslide-action-cleanup'. Similarly, with the hooks: 'ms-start-hook' sounds like it might have something to do with the 'ms-init/ms-end' methods. 'ms-begin-presentation-hook' or 'waterslide-present-hook', perhaps (if you chose M-x waterslide-present as your command). 'ms-narrow-hook' is probably fine, but 'ms-slide-hook' or 'waterslide-step-slide-hook' would be clearer and match with 'ms-after-last-slide-hook'. Consider aliasing or renaming 'ms-after-last-slide-hook' to 'ms-end-presentation-hook'. Also, I'm always torn on start/end vs. begin/end. The latter sounds better to me. Nearly nobody uses 'finish', though. Finally, thank you for the videos, they're fun and I love learning modern patterns for Emacs Lisp.
I like begin presentation and end presentation. One of the main name collisions is the method `ms-step-forward` and the command `ms-forward`. Maybe... `ms-presentation-foward`. for the command I like begin instead of init. Since begin and end are quite common in Elisp, it makes the role of final more obvious. Is final still ambiguous to likely users? Hard for me to guess. `attr_methods` is definitely wrong. I mainly export org to texinfo, so I forgot that these are supposed to be namespaced over the export backend. In this case, macro slides is kind of like an export backend. If you have more thoughts, catch up in a few days with an issue. A new draft of the naming should be in place by then.
Absolutely marvelous! In my daily work presentations and communication are very important and I was always looking for a way to make presentations that I would not be ashamed off from within emacs!
None other than Eric S Raymond recently appeared on the mailing list with results of investigating moving the runtime and compiler implementation over to SBCL. It's not the first time. Guile Emacs was another such effort. What's more critical across all of open source is enabling these projects, which have huge demand, to actually reach completion. That's the focus of Positron.
