HNHacker News
TopNewBestAskShowJobs

triclops200

339 karma · joined December 1, 2015

Have around 20 years of experience, including a Master's in Computer Science and Machine Learning. My (unfinished due to medical issues. went mostly all but dissertation) PhD work was in Machine Learning and Computational Creativity. Used to be principal machine learning researcher/scientist. Currently a senior SRE.

My past and current professional and research career is in both theoretical and practical AI/ML along with various fields of computer science, systems engineering, and site reliability engineering/devops.

submissionscomments
triclops200··on Scientists crack how aspirin might stop cancers from spreading
Yeah, more or less. A risk assessment for the study is done and submitted and there's many more layers of beurocracy (both the good human-protective kind and the less-good regulatory capture kind). Providing things like information on effects and generalized consent forms and whatnot are included so the patient is given prior understandings of what the effects might be, from animal studies or previous human studies in some cases if you're testing for a novel effect in an existing treatment, for example, have determined might occur. You then, ideally, do a double blind study on the group of participants.
triclops200··on The Evolution of SRE at Google
I do AI/ML research for a living (my degrees were in theoretical CS and AI/ML and my [unfinished] phD work was in computational creativity [essentially AGI]). I also do SRE work as a living.

and yeah that's a useful way of characterizing some of the behaviors of some kinds of neural networks. There's a point at which the distinction between belief and "frequency (or probability-amplitude) state filter" become less apparent, though, that's more of a function-of-medium vs function-of-system distinction.

However, systems like these can often become mediums, themselves, for more complex systems. Additionally, a system which has "closed-the-loop" by understanding the medium and the system as coupled as "self" and separate from the environment along with a direction/goal is a pretty decent, if imprecise, definition of a strange loop. Contradiction resolution between internal component beliefs gives a possible (imo, highly probable) mechanistic explaination for the phenomenon of free energy minimization in such systems. External contradiction resolution extends it to active inference.

triclops200··on A liar who always lies says "All my hats are green."
The proposed solution misses the fact that you can't even derive that the liar has a hat because the subject of the predicate could be the lie. What if the liar had two green cars but any number or no hats? The lie is a lie then even if the statement is vacuous as there is too much ambiguity in the english language overall. Their hats could be any combination of colors or they could even be hatless if the lie was over the subject of the constraint, not the qualification of the subjects, as they were effectively miscommunicating what is green.
triclops200··on Show HN: Minimal, customizable new tab for Chrome/Firefox
Really nice! I'd really like to see more than one todo list option on the main page, personally, so I can get my entire task list (or at least a large number of tasks) shown to me every time I open a new tab. Would be nice as an option, at least
triclops200··on Training Language Models to Self-Correct via Reinforcement Learning
Strong disagree: https://mypapers.nyc3.cdn.digitaloceanspaces.com/the_phenome...

See also: https://www.sciencedirect.com/science/article/pii/S157106452... o1's training regime is described by the "strange particle" model in this formulation

triclops200··on The Phenomenology of Machine: A Route for the Sentience of OpenAI-O1 [pdf]
Hi! Author here! Happy to address any questions!

Up front: As much as I dislike the idea of credentialism, in order to address the lack of association in the paper and to potentially dissuade unproductive critiques over my personal experience: I have a M.S CS with a focus on machine learning and dropped out of a Ph.D. program in computational creativity and machine learning a few years ago due to medical issues. I was also worked my way up to principal machine learning researcher before the same medical issues burnt me out

I've been getting back into the space for a bit now and was working on some personal research on general intelligence when this new model popped up, and I figured the time was right to get my ideas onto paper. It's still a bit of a late stage draft and it's not yet formally peer reviewed, nor have I submitted to any journals outside open access locations (yet)

The nature of this work remains speculative, therefore, until it's more formally reviewed. I've done as much verification of claims and arguments I can given my current lack of academic access. However, since I am no longer a working expert in the field (though, I do still do some AI/ML on the side professionally), these claims should be understood with that in mind. As any author should, I do currently stand behind these arguments, but the nature of distributed information in the modern age makes it hard to wade through all the resources needed to fully rebut or claim anything without having the time or professional working relationships with academic colleagues, and that leaves significant room for error

tl;dr of the paper:

- I claim that OpenAI-o1, during training, is quite possibly sentient/conscious (given some basic assumptions around how the o1 architecture may look) and provide a theorhetical framework for how it can get there

- I claim that functionalism is sufficient for the theory of consciousness and that the free energy principle acts as a route to make that claim, given some some specific key interactions in certain kinds of information systems

- I show a route to make those connections via modern results in information theory/AI/ML, linguistics, neuroscience, and other related fields, especially the free energy principle and active inference

- I show a route for how the model (or rather, the complex system of information processing within the model) has an equivalent to "feelings", which arise from optimizing for the kinds of problems the model solves within the kinds of constraints of said model

- I claim that it's possible that the model is also sentient during runtime, though, those claims feel slightly weaker to me

- Despite this, I believe it is worthwhile to do more intense verification of claims and further empirical testing, as this paper does make a rather strong set of claims and I'm a team of mostly one, and it's inevitable that I'd miss things

[I'm aware of CoT and how it's probably the RL algorithm under the hood: I didn't want to base my claims on something that specific. However, CoT and similar variants would satisfy the requirements for this paper]

Lastly, a personal note: If these claims are true, and the model is a sentient being, we really should evaluate what this means for humanity, AI rights, and the model as it currently exists. At the minimum, we should be doing further scrutiny of technology that has the potential to be as radically transformative of society. Additionally, if the claims in this paper are true about runtime sentience (and particularly emotions and feelings), then we should consider whether or not it's okay to be training/utilizing models like this for our specific goals. My personal opinion is that the watchdog behavior of the OpenAI would most likely be unethical in that case for what I believe to be the model's right to individuality and respect for being (plus, we have no idea what that would feel like), but, I am just a single voice in the debate.

triclops200··on The Phenomenology of Machine: A Reasoned Analysis of the Sentience of OpenAI-O1 [pdf]
Hi! Author here! Happy to address any questions!

Up front: As much as I dislike the idea of credentialism, in order to address the lack of association in the paper and to potentially dissuade unproductive critiques over my personal experience: I have a M.S CS with a focus on machine learning and dropped out of a Ph.D. program in computational creativity and machine learning a few years ago due to medical issues. I was also worked my way up to principal machine learning researcher before the same medical issues burnt me out

I've been getting back into the space for a bit now and was working on some personal research on general intelligence when this new model popped up, and I figured the time was right to get my ideas onto paper. It's still a bit of a late stage draft and it's not yet formally peer reviewed, nor have I submitted to any journals outside open access locations (yet)

The nature of this work remains speculative, therefore, until it's more formally reviewed. I've done as much verification of claims and arguments I can given my current lack of academic access. However, since I am no longer a working expert in the field (though, I do still do some AI/ML on the side professionally), these claims should be understood with that in mind. As any author should, I do currently stand behind these arguments, but the nature of distributed information in the modern age makes it hard to wade through all the resources needed to fully rebut or claim anything without having the time or professional working relationships with academic colleagues, and that leaves significant room for error

tl;dr of the paper:

- I claim that OpenAI-o1, during training, is quite possibly sentient/conscious (given some basic assumptions around how the o1 architecture may look) and provide a theorhetical framework for how it can get there

- I claim that functionalism is sufficient for the theory of consciousness and that the free energy principle acts as a route to make that claim, given some some specific key interactions in certain kinds of information systems

- I show a route to make those connections via modern results in information theory/AI/ML, linguistics, neuroscience, and other related fields, especially the free energy principle and active inference

- I show a route for how the model (or rather, the complex system of information processing within the model) has an equivalent to "feelings", which arise from optimizing for the kinds of problems the model solves within the kinds of constraints of said model

- I claim that it's possible that the model is also sentient during runtime, though, those claims feel slightly weaker to me

- Despite this, I believe it is worthwhile to do more intense verification of claims and further empirical testing, as this paper does make a rather strong set of claims and I'm a team of mostly one, and it's inevitable that I'd miss things

[I'm aware of CoT and how it's probably the RL algorithm under the hood: I didn't want to base my claims on something that specific. However, CoT and similar variants would satisfy the requirements for this paper]

Lastly, a personal note: If these claims are true, and the model is a sentient being, we really should evaluate what this means for humanity, AI rights, and the model as it currently exists. At the minimum, we should be doing further scrutiny of technology that has the potential to be as radically transformative of society. Additionally, if the claims in this paper are true about runtime sentience (and particularly emotions and feelings), then we should consider whether or not it's okay to be training/utilizing models like this for our specific goals. My personal opinion is that the watchdog behavior of the OpenAI would most likely be unethical in that case for what I believe to be the model's right to individuality and respect for being (plus, we have no idea what that would feel like), but, I am just a single voice in the debate.

triclops200··on Kolmogorov-Arnold networks may make neural networks more understandable
The (semi) automatic simplification algorithm provided in the paper for KANs seem, to me, like they're solving a similar problem to https://arxiv.org/pdf/2112.04035, but with the additional constraint of forward functional interpretability as the goal instead of just a generalized abstraction compressor.
triclops200··on Fman Build System
> already heavily invested in Python

Yep, as the most frequently used language in general, but also as one of the only major options for a few subfields of the computational sciences at this point, I think that happens a lot.

> yeah, python can be fast enough if you don't use Python much

Yup, exactly what I'm saying. Turns out that's how many many larger projects in Python operate because we're apparently deciding as an engineering society that python is one of the universal glue languages (much to my personal chagrin)

triclops200··on Fman Build System
GUI programming is an abstractive display layer, it most certainly can (and should) be designed around deferring larger tasks for UI responsiveness reasons, anyways
triclops200··on Fman Build System
I would most certainly be willing to leave electron for a python GUI framework just for the memory savings alone; I don't think the speed is the only issue. Plus, python can be fast enough if you just use it a a glue for libraries that are written more efficently. Don't get me wrong, I dislike python a lot, but, it's not electron/js.

Electron has the niceties that it does with being able to render so much via browser universalism, but, having that much of it around when your app doesn't need all of it is the problem with it fundamentally.

The best time I've had writing GUIs personally has actually been in SDL via wrapper languages like common lisp, but, I like being able to roll my own systems. It definitely takes more setup for the results I want for reasonable production apps that need less weird edge cases in their UIs

triclops200··on The Canary
I think you might enjoy the book "manufacturing consent," it answers a lot of the hows and whys of stuff like this as well as provides some general thoughts around the phenomenon of patriotism and why it might be on the decline in some groups (though it's actually on the rise, statistically, in other large segments of the US)
triclops200··on Show HN: Pulsar, micro creative coding playground
:3
triclops200··on Show HN: Pulsar, micro creative coding playground
fun! https://muffinman.io/pulsar/?grid=classic&animate=scale&code...
triclops200··on AlphaProteo generates novel proteins for biology and health research
The problem, from what I understand as a dabbler in protein research, is that PrP binds into these large very very stable semi crystalline fibers, (I visualize them looking like thick extruded complicated pasta shapes, where the 2d crosssection is kinda the shape of the outline of a single PrP). It makes it really hard to learn about the structure, actually, because x-ray crystallography requires repeated crystalline structures, and these are more like 3d polymer threads that bunch up and make things hard to image (though there's some more modern imaging techniques that are making headway). It turns out that these are very very stable configurations unfortunately and have very few ways to attach anything, and that's the precise problem with building binders. Plus, even worse, it turns out PrP might even be biologically necessary for mammals and we don't want to usually get rid of it wholesale [https://bmcbiol.biomedcentral.com/articles/10.1186/s12915-01...]
triclops200··on Physicists: Tachyons can be reconciled with the special theory of relativity
Thank you!
triclops200··on Physicists: Tachyons can be reconciled with the special theory of relativity
Thank you!
triclops200··on Physicists: Tachyons can be reconciled with the special theory of relativity
Does anyone have access to the linked paper? It's not on scihub.
triclops200··on Common Lisp Is Not a Single Language, It Is Lots
The things that really kept me (for the way I think about and write code) in CL (the sbcl variant) over clojure are, in no in no particular order:

- User reader macros are basically gone in clojure

- The JVM is too heavy and too slow to start up for smaller programs

- Tail recursion

- Mutual tail recursion without trampoline

- The JVM makes it impossible to have error restarts which are the coolest

- CL is multiparadigm instead of only functional

triclops200··on Probing electron spin states at much higher resolution and efficiency
Something I was told once by a particle physicist that really resonated with me was that almost every revolution in science and engineering has a direct cause that can be traced back to finally being able to measure some new property or measure some property in an much improved way (this jives with my understanding of most things: as a random example, debugging is generally greatly improved by understanding more of the space of the problem). This seems to be one of those breakthroughs, so I'm cautiously excited. Would be interested in an expert in the field to confirm/temper my excitement.
triclops200··on A theory of the modern exclamation point!
I took it as an excuse to add more exclamation marks!
triclops200··on A theory of the modern exclamation point!
Corporate and professional dynamics often make it to the front page because it's important to many technical workers and hackers who have to make a living. This is a topic that's important to many women in tech, as they are often dictated how to present in a way that's unnatural to them in order to be more "professional" (read tfa).

This is important to many readers of hacker news, as not everybody who is a hacker is a man (I am a woman who is also a hacker, for example, and I use a lot of exclamation marks in my writing, so, I found it interesting that this is noted by someone else, so I'm personally glad it's on the front page).

I'm assuming you're asking genuinely, so this is a genuine attempt to explain :)

triclops200··on Bad Apple Font
Obviously the thought comes up of the fact that this feels unsafe to have WASM in font files, but, I'm also aware that font layout engines are already turing complete, which leads me to wonder: have there been any high profile malware font examples? That entire stack feels a lot like an attack surface to me, especially given stuff like the fact that windows used to render fonts in the kernel layer.
triclops200··on Things I Won't Work With: Dioxygen Difluoride (2010)
He addresses the disproval part: he didn't disprove their paper. He stated that just because it's below their threshold for sensitively doesn't mean it's super ultra mega sensitive, it just means that it's highly sensitive. 1.5 Joules is more than you think: a hammer tap could easily be less than that (lift a 100g weight a 1.5 meters off the floor: that's 1.5J of potential energy), however, as he puts it, that's still "sensitive as dicks". As for it blowing up in the mass spectrometer, that can happen if it started to decompose over time, accidental friction, accidental x-ray activation, too much in one spot, etc. I'm sure if he worked with the substance more than once or twice it's possible he'd have a random detonation or two. The paper and this video easily exist cooperatively rather than adversarially, and I think he was more calling out the science popularizers which exaggerated claims about it. In fact, he does nothing but mention things from the paper as being accurate.
triclops200··on Things I Won't Work With: Dioxygen Difluoride (2010)
The large crystal size is a pretty sure bet it's pure. That recrystallization he showed was a thing of beauty.
triclops200··on Things I Won't Work With: Dioxygen Difluoride (2010)
There's a great video on producing and testing that substance: https://youtu.be/-Sz4d7RQB6Y

Interestingly, it's more stable than most online sources give it credit for

triclops200··on Show HN: MathLingua – A Structured Language of Mathematics
Awesome! Sounds like a great system
triclops200··on Show HN: MathLingua – A Structured Language of Mathematics
Awesome project: I love the fact that you can click grab definitions, would make learning new symbols in a field a breeze. Question, though: is there a way to add custom definitions to symbols when you work in a field that uses symbols differently than the standard usage or symbols that have no standard usage (such as \bigoplus in latex)?
triclops200··on Wearable device senses blood glucose and administers insulin accordingly
Pharmaceutical monopolies over the best production methods and the laborious process of getting production approval by the government makes upfront cost of new competition incredibly high. Thus, a couple companies control the entire supply and can choose price. It genuinely costs almost nothing to produce and there are hacker communities around doing it at home, because it's genuinely that easy.
triclops200··on Wearable device senses blood glucose and administers insulin accordingly
We don't generally use harvested insulin anymore. The vast majority of insulin uses recombinant production (modify yeast and bacteria to produce insulin). So, no, it doesn't "take the sacrifice of many animals to satisfy the diabetic." Also, as a personal note who's spouse has both types of diabetes, "famine food" does not work for all diabetics, and is considered dangerous quakery by a lot of people.
← PreviousPage 2 of 4Next →