I was also recently impressed by Zhaos "Mathematical Foundation of Reinforcement Learning", free textbook and video lectures on YT : https://github.com/MathFoundationRL/Book-Mathematical-Founda...
If you dont have a lot of time, at least glance at Zhaos overview contents diagram, its a good conceptual map of the whole field, imo .. here :
https://github.com/MathFoundationRL/Book-Mathematical-Founda...
and maybe watch the intro video.
[1] https://scholar.google.com/scholar?start=10&q=andreas+krause...
Long time ago at uni we studied these things for a while.. and even made a Prolog interpreter having both F+IF (probability + confidence) coefficients for each and every term..
Likewise in your metric, if all answers are the same despite perturbations then it's more likely to be ... true?
I'd really like to see a plot of your metric versus the SimpleQA hallucation benchmark that OpenAI uses.
Is this helpful?
https://en.wikipedia.org/wiki/Calibration_(statistics)
Example: Efficient and Effective Uncertainty Quantification for LLMs (https://openreview.net/forum?id=QKRLH57ATT)
Simpler example: represent an LLM as a green field with objects, where humans are the only agents:
You stand near a monkey, see chewing mouth nearby, go there (your prompt now is “monkey chews”), close by you see an arrow pointing at a banana, father away an arrow points at an apple, very far away at the horizon an arrow points at a tire (monkeys rarely chew tires).
So things close by are more likely tokens, things far away are less likely, you see all of them at once (maybe you’re on top of a hill to see farther). This way we can make a form of static place AI, where humans are the only agents
My mind turned into an infinitely large department store where each aisle was a concurrent branch of thought, and the common ingredient lists above each aisle were populated with words, feelings and concepts related to each branch.
The PA system replaced my internal monologue, which I no longer had, but instead I was hearing my thoughts externally as if they were another person's.
I was able to walk through these aisles and marvel at the immense, fractal, interdependent web of concurrent thought my brain was producing in realtime.
Many of these directives are contradictory. The coexistence of these contradictory programs is what we call inner conflict. This conflict causes us to constantly check ourselves while we are caught in the opposition of polarity. Another metaphor would be like a computer with many programs running simultaneously. The more programs that are running, the slower the computer functions. This is a problem then. With all the programs running that are demanded of our consciousness in this modern world, we have problems finding deep integration.
To complicate matters, the programs are reinforced by fear. Fear separates, love integrates. We find ourselves drawn to love and unity, but afraid to make the leap.
What I found to be the genius of LSD is that it really gets you high, higher than the programs, higher than the walls that mask and blind one to the energy destroying presence of many contradictory but hidden programs. When LSD is used intentionally it enables you to see all the tracks laid down, to explore each one intensely. It also allows you to see the many parallel and redundant programs as well as the contradictory ones.
It allows you to see the underlying unity of all opposites in the magic play of existence. This allows you to edit these programs and recreate superior programs that give you the insight to shake loose the restrictions and conflicts programmed into each one of us by our parents, our religion, our early education, and by society as a whole.”
~ Nick Sand, 2001, Mind States conference, quoted in Casey Hardison's obituary
The guy who’ll make the GUI for LLMs is the next Jobs/Gates/Musk and Nobel Prize Winner (I think it’ll solve alignment by having millions of eyes on the internals of LLMs), because computers became popular only after the OS with a GUI appeared. I recently shared how one of its “apps” possibly can look: https://news.ycombinator.com/item?id=43319726
Maybe this is why tokens and language are so useful for humans ? they might be the closest analog we have.
I wonder it this one might be a bit more accessible, although I guess the R/Python examples are helpful on the latter.
and frankly i would not recommend islr anymore today, too dated
Potential exists, probability is a mathematical description of its distribution. Every attribute is a dimension (vector). State is merely a passing measurement of resolve. Potential interacts through constructive and destructive interference. Constructive and destructive interference resolve to state in a momentary measure of “now” (an inevitability decaying proposition.)
Existential Reality is potential distributing, not arrangements of state.
Handy if you want help breaking this down.
'Laplace Approximation is a "quick and dirty" way to turn a complex probability distribution into a simple Gaussian (bell curve).
It works by finding the highest point (mode) and matching the curvature at that point.
It's fast and easy, but it can be very inaccurate and overconfident if the true distribution doesn't look like a bell curve.'
This document is the lecture notes.
Though I personally prefer to read these sorts of books directly from pdf, and am grateful to them for sharing it on arxiv.
Also it should use LLMs and the blockchain.
But this would be nice there are a number of papers and such that if you could submit an arXiv link to a print service I would probably buy a copy. I wonder why no one does it.
Books have exercises. It's your job to engage.
This book, in particular, has 3 pages of Problems per chapter. The only way to learn the math is to do all of them.
Current gen of LLM programming AIs might make it less leg-work to make these
> open article
> "holy shit it's 400 pages"
> realize i already have a grasp on most of the material from school
> "phew"
> oh this stuff is cool, just like i remember...
> proceed to read all 400 pages
well done! :clap:
My guess is that most of the content in the book is several years old (it's apparently based on an ETH Zurich class), despite the PDF being compiled this year, which would explain why it doesn't cover the state of the art.
