The labs
Real mathematics you can run: Bayes’ theorem, variational free energy, predictive coding and the POMDP machinery behind every UNI agent — each one an interactive laboratory, not a slideshow.
Read this first. Three kinds of entry are listed here and they are drawn differently on purpose. Run them here are labs served from this site, byte-verified on every build against a pinned commit of the public Workbench repository. Live on the campus are labs on another host: the links are declared destinations, never probed, and this page does not claim they answer right now. Built, not yet hosted are labs that exist and are served from nowhere — they are listed with what is missing rather than dropped. The labs are MIT-licensed and come from the public Workbench repository. Every lab served here runs entirely in your browser: no account, no server, no analytics, CPU only.
7 served from this site · 6 live on the campus · 1 built but hosted nowhere · read at 2026-08-31T07:18:53.047Z
These labs are the practical half of a curriculum: the course is the syllabus they teach inside, and the hallway places them among the rest of the estate.
Walk in
The laboratory’s own rooms, served whole from this site — 6 rooms and 25assets, every file an exact copy of a named path at a named commit, verified byte-for-byte on every build. The classroom is also the site’s front door.
The room you stand in: a photo-real laboratory with pixel-measured, hash-bound hotspots — the whiteboard, the specimen bench, the frequency wall and the mind-body wall.
the two viewer-API fetches fail closed into the page's own fallbacks; everything else — the room, the anchors, the specimens, the microscopy — is served from this site
Every model the laboratory carries, as the classroom's own model wall presents them.
the live model-registry fetch fails closed; the page's built-in content is served as committed
How the laboratory is built: the world process, the recorder, the reconstruction and the agent, kept deliberately separate.
fully static — no fetches
The dated state of the parity programme: the ladder, the gates, and what stands where — as of the date in its own title.
a dated document by design; it says its own date
Every matrix drawn, every step of the mathematics shown as it happens — the colony's generative model on one board.
fully static — no fetches; the defaults on the board are the committed defaults
The organism view of the same mathematics: the factors as organs of one living loop.
fully static — no fetches
Run them here
These 7 labs are copies of files from the public Workbench repository, pinned at commit c2c9c7246251. On every build of this site, each served copy is hashed and compared byte-for-byte against the same file at that pinned commit; a mismatch fails the build rather than shipping. That check is what ties the lab you run to the real code — the cite line under each card carries the hash so you can rerun the comparison yourself. (At the last read, the public repository’s head was the pinned commit itself.)
Bayes' theorem as a physical machine: prior chips pour through likelihood gates and what lands in the tray is the posterior.
- The math
- P(h|e) = P(e|h)P(h) / P(e), run as countable chips rather than symbols
- Time
- 10–20 minutes
Open the lab — a single HTML file, served from this site, runs in your browser.
byte-verified against
learninglabs/BayesChips.html @ c2c9c7246251 — sha256
61e4db4f88baf036 —
sourceThe discrete-time generative model behind every UNI agent: A, B, C, D matrices you can set by hand, then step and watch beliefs move.
- The math
- Hidden-state inference and expected free energy over policies in a partially observed Markov decision process
- Time
- 15–30 minutes
Open the lab — a single HTML file, served from this site, runs in your browser.
byte-verified against
learninglabs/active_inference_pomdp_machine.html @ c2c9c7246251 — sha256
e0d80c4bd573aac8 —
sourceEquation 4.19 taken apart on a bench: every term of variational free energy shown, moved, and re-assembled.
- The math
- F = E_q[ln q(s) − ln p(o,s)] decomposed into energy and entropy, then into accuracy and complexity
- Time
- 15–30 minutes
Open the lab — a single HTML file, served from this site, runs in your browser.
byte-verified against
learninglabs/free_energy_forge_eq419.html @ c2c9c7246251 — sha256
ef0899cd8d45262b —
sourceA predictive-coding hierarchy built floor by floor: prediction errors climb, predictions descend, precision decides who is heard.
- The math
- Hierarchical Gaussian generative models under the Laplace approximation — precision-weighted prediction error
- Time
- 15–30 minutes
Open the lab — a single HTML file, served from this site, runs in your browser.
byte-verified against
learninglabs/laplace_tower_predictive_coding_builder.html @ c2c9c7246251 — sha256
1dda273ce562657e —
sourceThe gentlest start here: one frog, two lily pads, and the whole idea of a generative model before any matrix is named.
- The math
- A two-state, two-observation generative model — the smallest POMDP that can still surprise you
- Time
- 8–15 minutes
Open the lab — a single HTML file, served from this site, runs in your browser.
byte-verified against
learninglabs/jumping_frog_generative_model_lab.html @ c2c9c7246251 — sha256
5abad46eba4e167e —
sourceThe brain-shaped map: cortical circuits, predictive coding and neuromodulation located on one navigable atlas.
- The math
- Message passing between cortical layers as belief propagation; neuromodulators as precision control
- Time
- 15–30 minutes
Open the lab — a single HTML file, served from this site, runs in your browser.
byte-verified against
learninglabs/active_inference_atlas_educational_sim.html @ c2c9c7246251 — sha256
8ebe6c2214253b66 —
sourceFigure 5.5 as a working studio: the full perception–action cycle traced through a living diagram.
- The math
- The perception–action loop of active inference, factor by factor
- Time
- 15–30 minutes
Open the lab — a single HTML file, served from this site, runs in your browser.
byte-verified against
learninglabs/anatomy_of_inference_studio.html @ c2c9c7246251 — sha256
3dcb75e034139444 —
sourceLive on the campus
These 6 labs run on the campus site, not here. Every link below is a declared destination, not a probe — this page is a static export and cannot honestly claim what is answering right now.
Three precision dials — trust in the senses, trust in actions, commitment to a plan — steer a maze agent through visibly different behavioural regimes.
- The math
- Power-law precision on A and B matrices; policy temperature in π = σ(−G/T); EFE split into epistemic and pragmatic parts
NOT PROBED — the URL is a declared destination; a static build cannot honestly claim what answers right now
developed in the operator's campus repository (private); the engine is a verified translation of the Workbench's Elixir POMDP engine — no source link is offered, because the repository it lives in is not public.
Active inference with echolocation: sensing as an act the agent chooses to spend.
- The math
- Epistemic value of an active sensing action vs. its pragmatic cost
NOT PROBED — the URL is a declared destination; a static build cannot honestly claim what answers right now
operator's campus repository (private) — no source link is offered, because the repository it lives in is not public.
The math of stuck patterns: how a loop forms, holds, and breaks.
- The math
- Bifurcation structure of a belief–action loop under changing precision
NOT PROBED — the URL is a declared destination; a static build cannot honestly claim what answers right now
operator's campus repository (private) — no source link is offered, because the repository it lives in is not public.
The long-term cardio-renal loop as a control problem the body is always solving.
- The math
- Slow homeostatic regulation as inference over interoceptive states
NOT PROBED — the URL is a declared destination; a static build cannot honestly claim what answers right now
operator's campus repository (private) — no source link is offered, because the repository it lives in is not public.
An autopoietic service cell under active-inference test: a system that maintains itself or stops existing.
- The math
- Autopoiesis read as existence-preserving free-energy minimisation
NOT PROBED — the URL is a declared destination; a static build cannot honestly claim what answers right now
operator's campus repository (private) — no source link is offered, because the repository it lives in is not public.
A program learning to read, one letter at a time — watched live as it happens.
- The math
- Sequential inference over character streams
NOT PROBED — the URL is a declared destination; a static build cannot honestly claim what answers right now
operator's campus repository (private) — no source link is offered, because the repository it lives in is not public.
Built, not yet hosted
These exist and are served from nowhere. They are printed rather than dropped, because a quietly shortened list is how an inventory starts lying.
The Bayes Museum — 50 exhibits
not hosted state as declared: built, verified self-contained, not yet deployed
a Next.js application that does not yet set output:'export'; its 50 exhibits prerender cleanly and carry no external calls, but the repository has no public remote, so there is nothing a citation could resolve to yet
What would change this: a public repository plus a one-line static-export config
Wondering how this connects to the rest of the estate? The hallway has one door per project, and the course walks these labs in teaching order.