80topics
Game Theory Chat
Strategy, equilibrium and incentives, from the prisoner’s dilemma to mechanism design.
FIG. 01 / Game theory
Free · no account · open source
Study a concept. See the steps.
For students and self-directed learners: explore game theory, data science, reinforcement learning, linear algebra, the web platform and the Islamic Golden Age. Ask an assistant about a covered topic, read an example or try an interactive tool.
No account or payment needed. Typed chat questions are processed on your device; see coverage and limits.
FIG. 00 / THE ACTIVATION PLOTf(x) = max(0, x)
max(0, x), is the function this site is named after: silent while the input is negative, then straight up. Push x right and each subject switches on.Pick a subject
Six assistants, each with its own curated knowledge base
50topics
Golden Age Inquiry
Scholars, institutions and ideas of the Islamic Golden Age: algebra, optics, medicine and philosophy.
Try askingAl-KhwarizmiHouse of WisdomIbn al-Haytham (Alhazen)
FIG. 02 / Islamic Golden Age
41topics
Data Science Chat
Statistics, evaluation and machine-learning concepts with the math shown.
Try askingTrain/Test SplitLogistic RegressionProbability Distributions
FIG. 03 / Data science
16topics
Reinforcement Learning Chat
Rewards, value functions, bandits, Q-learning and policy gradients.
FIG. 04 / Reinforcement learning
16topics
Linear Algebra Chat
Vectors, matrices, eigenvalues and projections: the algebra behind machine learning.
Try askingEigenvalues and eigenvectorsDot productLeast squares
FIG. 05 / Linear algebra
16topics
Web Platform Chat
Browser JavaScript, HTTP caching, WebAssembly, workers and local storage.
Try askingService workerEvent loopWebAssembly
FIG. 06 / Web platform
Assistants retrieve and assemble curated fragments. They can misunderstand a question, do not execute code and cannot solve arbitrary exercises.
How it works
From your question to an answer, one step at a time
INSIDE YOUR BROWSER
01 Retrieve
02 Read signals
03 Decide
04 Compose
05 Render
- 1
Retrieve
Keyword matching and field-weighted BM25 find candidate topics. An optional ~22 MB sentence model adds sentence-vector comparisons.
- 2
Read the signals
Entities, intent cues and ranked scores merge into one structured packet, all on your device.
- 3
Decide
A 13,079-parameter policy network chooses how to respond: answer, clarify, compare or follow up.
- 4
Compose
The assistant assembles an explanation from curated fragments. It can choose the wrong topic or miss what you meant.
- 5
Render
KaTeX typesets the math and session memory keeps follow-ups in context.
Interactive tools
Change a number, watch the algorithm respond
- Neural Network ExplorerAdjust weights and activations and watch a forward pass move through every neuron.→
- Activation Functions ExplorerPlot ReLU, Sigmoid, Tanh, GELU and more with their derivatives, and see why slope decides whether learning continues.→
- Gradient Descent LabClick a starting point, set the learning rate and follow each step down a loss surface.→
- Backpropagation VisualizerWalk the chain rule backwards through a 2→2→1 network with every gradient derived.→
- K-Means Clustering PlaygroundAdd points, choose k and step through assignment and centroid updates as inertia shrinks.→
- Decision Tree ExplorerGrow a tree one split at a time and watch Gini impurity fall, then overfitting appear.→
Worked examples
Latest guides
- Principal Component Analysis Explained: Finding Directions of Most VariancePrincipal component analysis rotates data onto perpendicular axes ordered by variance. Learn how the components are found, the role of scaling, and how many to keep.
- Eigenvalues and Eigenvectors Explained: Directions a Matrix Only StretchesAn eigenvector keeps its direction when a matrix acts on it, and the eigenvalue is the scaling factor. Learn the idea…
- Greedy vs Beam Search DecodingGreedy decoding is fast but myopic; beam search explores alternatives at k times the cost. Learn both, plus sampling…
From the makers
Go further than the site
Book · 224 pages
How to Remain Valuable When Intelligence Becomes Cheap
A 224-page practical guide to the human, economic and strategic advantages that stay valuable as AI takes on more of the world's cognitive work.
Toolkit · Separate Gumroad product
On-Device Chatbot Builder Toolkit
If you want to build a related system, the toolkit offers frameworks, templates and evaluation guides for the stages from input to response. Open its Gumroad listing to check the contents and current purchase terms.
View toolkit on Gumroad
Ebook · Separate Gumroad product
Fringe Learning: Resource-Efficient RL for Edge ML
Build your own custom AI systems with practical reinforcement learning methods designed for resource-constrained edge machine learning.
View ebook on GumroadAbout
An open-source experiment in simpler chatbots
Yunus Emre Vurgun builds ReLU.chat as a learning resource and an inspectable experiment in browser-based retrieval. You can read the knowledge pages, try the assistants and review the source without an account.
The system combines keyword retrieval, an optional sentence-matching model, a small response policy and curated explanation fragments. These methods make the process visible; they do not guarantee a correct answer. Hosting, optional downloads, external links and separate subscriptions or purchases use the network.