Patent Pending — Bee Tree Holdings LLC

What if computers could wire themselves?

The Universal Constraint Engine defines simple rules about what can and can't happen. The computer figures out the rest — memory, logic, learning — all by itself. No programming required.


The Core Idea

Think of It Like DNA

DNA doesn't contain instructions for every possible behavior. It contains simple rules. Those rules interact, and complex life emerges — billions of cell types, organs, brains, all from the same genetic rules. UCE works the same way for computing. We define simple rules about what CAN and CAN'T happen. The computer figures out the rest — memory cells, logic gates, oscillators — all by discovering them automatically from the constraints.

Traditional Neuromorphic Chips

1 Engineer designs each neuron
2 Programs the behavior explicitly
3 Gets exactly what they programmed

UCE Approach

1 Engineer defines simple rules
2 Engine discovers emergent behaviors
3 Discovers things nobody programmed

See It In Action

Interactive Constraint Explorer

Click "Start Walkthrough" to see how constraints create computational behaviors step by step. Or toggle individual constraints below to experiment.

Constraint Network

Step 1: Seven Switches

The system has 7 binary quantities: E, I, T, F, R, L, P. Each is either ON or OFF. This is the building block.

Step 2: Add a Rule

Rule: "E can only turn on when L is on" — this is a constraint. When L is off, E is locked. This gates the excitation.

Step 3: Chain Rules Together

Add more constraints: E gates T, T gates F, F gates P. Each constraint depends on the previous. The system now has a pathway.

Step 4: Inhibitory Path

Add another rule: "I and E can never both be on." This creates mutual exclusion. The system can now make decisions.

Step 5: Behaviors Emerge

The constraints lock into 45 valid states. Within those states, it discovers memory (each quantity remembers), logic gates (cascading pathways), oscillators (self-sustaining cycles), and stable attractors.

Toggle Constraints


The Problem

Neuromorphic Chips Hit a Wall

Every major neuromorphic chip on the market today requires engineers to hand-code neuron behavior. That's like building a brain one cell at a time.

⚙️

Hand-Crafted Neurons

Engineers design each neuron's behavior before fabrication. The computational properties are baked in, not emergent. Want a different neuron? Redesign it.

📊

Transistor Bloat

1,000-2,500 transistors per neuron. A million-neuron network needs billions of transistors. That limits density and drives up power and cost.

🔍

No Discovery Method

There's no systematic way to find which designs produce which computational behaviors. It's all trial and error. There was no framework — until now.


The Solution

Define Rules. Discover Behaviors.

Instead of designing neurons, you declare constraints. The UCE automatically discovers every emergent behavior — without writing a single line of neuron code.

7
Conserved Quantities
18
Constraint Rules
45
Valid System States
18
Emergent Behaviors

The Core Design

Seven Quantities. Infinite Complexity.

The Constraint-Driven Self-Emergent Neuron (CSEN) has seven binary quantities. Each maps to a biological neural function. Their interactions produce all computational behavior.

E Excitatory Input
I Inhibitory Input
T Threshold
F Fire / Output
R Refractory Period
L Leak / Decay
P Potentiation / Learning

How Behaviors Emerge

Three Types of Rules

All computational richness comes from just three types of constraints applied to the seven quantities.

Allow Rules

6 rules — AND conjunction

Define when a quantity may change. Multiple conditions must ALL be true. Create the gating network that chains quantities together into pathways.

Forbid Rules

8 rules — OR disjunction

Define when a quantity cannot change. ANY satisfied condition blocks the toggle. Create inhibitory constraints and safety locks.

Mutual Exclusion

4 rules — exclusive-or

Eliminate states where two quantities are both active. Create phase opposition (excitation vs. inhibition, firing vs. refractory).


What Emerges Automatically

18 Behaviors. Zero Programming.

The engine processes the 18 constraint rules over 7 quantities and discovers 18 distinct computational behaviors across four categories.

It Remembers Things

All seven quantities retain their value unless specific gating conditions unlock them. Each acts as an independent binary memory cell — 7 bits of per-neuron storage, with no flip-flops required.

It Makes Decisions

A dual-pathway cascade emerges: excitation (L→E→T→F→P) and inhibition (L→R→I). Leak (L) acts as the master control for the entire neuron. Gating creates logic operations.

It Creates Rhythms

Four attractor cycles emerge as oscillatory patterns: potentiation oscillator, sensitization cycle, and more. These are self-sustaining excitation-learning feedback loops.

It Finds Stable States

Nine stable equilibrium states form an attractor basin. The neuron naturally settles into predictable resting configurations, providing the foundation for reliable computation.

Why This Matters

These behaviors are not programmed. They emerge from the constraint rules. This means the CSEN can discover computational properties that engineers never explicitly designed. Traditional neuromorphic chips do the opposite — every behavior must be hand-coded. UCE flips the paradigm: define rules, discover behaviors.


How We Compare

Not an Incremental Improvement

UCE is a fundamentally different architecture. Here's how it stacks against the state of the art.

Feature Conventional Chip A Conventional Chip B CSEN (Ours)
Neuron Design Hand-programmed Fixed LIF model Emergent from constraints
Transistors/Neuron ~2,300 ~5,400 Target: < 100
Per-Neuron Memory Configurable Limited 7 bits intrinsic
Behavior Discovery Manual design Manual design Automatic enumeration
Learning Rule Programmable STDP External Emergent (potentiation)
Memory Scaling Linear Fixed 3N-1 bits (N neurons)
How It Works Engineer programs each neuron Engineer programs each neuron Engineer defines rules, behaviors emerge

Intellectual Property

Protected Innovation

Patent applications filed with the United States Patent and Trademark Office, establishing priority dates for the core UCE technology stack.

Patent Pending

Universal Constraint Engine

Application No. 64/036,854

System and method for discovering emergent behaviors from declarative constraint rules over conserved quantities.

Patent Pending

CSEN Neuromorphic Architecture

Application No. 64/039,741

Constraint-driven self-emergent neuron architecture for neuromorphic computing with 7-quantity, 18-rule design.


Published Research

Read the Paper

Our technical paper presents the UCE architecture, worked examples, and implications for the future of neuromorphic computing.

📄

The Universal Constraint Engine: Emergent Neuromorphic Architectures from Declarative Constraint Rules

Stephen C. Kinney — Bee Tree Holdings LLC
April 2026  |  6 pages  |  CC BY-NC-ND 4.0
Constraint satisfaction • Neuromorphic computing • Emergent behavior • Hardware-agnostic architecture

The Future Computes Itself

Join the early access list. Be among the first to experiment with constraint-driven neuromorphic computing.