Core Concepts

Term	Definition	See Also
Hypervector	A high-dimensional binary vector (default 32,768 bits) that represents a concept, fact, or relationship. The fundamental data structure of AGISystem2.	Theory
Geometry	The number of dimensions (bits) in a hypervector. Common values: 1024, 8192, 32768, 65536. Higher geometry = more capacity but more memory.	API
Atom	A named hypervector representing a primitive concept (e.g., "Person", "loves", "John"). Atoms are the building blocks of knowledge.	Syntax
Fact	A bound hypervector encoding a relationship (e.g., "John loves Mary"). Created by binding atoms with position markers.	Syntax
Theory	A named collection of atoms, macros, and rules that defines a domain of knowledge. Theories can be loaded and composed.	Syntax
Session	An isolated reasoning context with its own scope, knowledge base, and vocabulary. Sessions are independent and can run in parallel.	API
Knowledge Base (KB)	A bundled hypervector containing all facts learned in a session. Used for query resolution via unbinding.	Theory
Vocabulary	The set of all known atoms in a session. Used for similarity matching when resolving queries.	API

Operations

Term	Definition	Properties
Bind (XOR)	Bitwise XOR of two vectors. Creates associations between concepts. Used for building facts.	Commutative, Associative, Self-inverse, Reversible
Bundle (Majority)	Bitwise majority vote across multiple vectors. Creates superpositions. Used for building the KB.	Result similar to all inputs, Not reversible, Capacity-limited
Similarity	Normalized Hamming distance between vectors. Returns 0.0-1.0. Used for matching and retrieval.	1.0=identical, 0.5=random, 0.0=inverse
Unbind	XOR with a partial query to extract unknown components from a bundled KB.	Same as Bind (XOR is self-inverse)

DSL Terms

Term	Syntax	Description
Destination	@name	Variable to store result
Export	@name:export	Store and export with name
Discard	@_	Execute but don't store
Reference	$name	Access existing variable
Hole	?name	Unknown value to find
Position Vector	Pos1...Pos20	Markers for argument order
Macro	macro...end	Reusable statement template

Reasoning Terms

Term	Definition
Query	A statement with holes that retrieves matching information from the KB.
Proof	A tree of reasoning steps that derives a goal from known facts and rules.
Backward Chaining	Proving a goal by recursively proving the premises of applicable rules.
Rule	An implication (If P then Q) used for inference.
Confidence	A score (0.0-1.0) indicating the reliability of a result.
Binding	The association of a query hole to its answer value.

Output Terms

Term	Definition
Decode	Extract operator and arguments from a vector using similarity search.
Summarize	Generate concise natural language from a decoded structure using templates.
Elaborate	Generate detailed narrative, optionally refined by LLM for fluency.
Phrasing Template	A pattern like "{Pos1} loves {Pos2}" for generating text from structures.
DSL Trace	A replayable script recording all operations for audit and verification.

Architectural Terms

Term	Definition
Core HDC Layer	Lowest layer: vector storage, Bind, Bundle, Similarity, ASCII stamp initialization.
Runtime Layer	State management: sessions, scopes, theory registry, KB, vocabulary.
Processing Layer	Logic: parser, executor, query engine, proof engine, decoder.
API Layer	Public interface: learn(), query(), prove(), summarize(), elaborate().
Scope	Variable namespace with hierarchy: Global > Theory > Session > Macro.
ASCII Stamp	Deterministic vector initialization from name strings for reproducibility.

Thresholds Reference

Threshold	Default	Meaning
Strong Match	> 0.80	High confidence, trust result
Good Match	0.65 - 0.80	Solid match, probably correct
Weak Match	0.55 - 0.65	Marginal, verify if critical
No Match	< 0.55	Near random, don't trust
KB Warning	100 facts	Approaching capacity limits
KB Critical	200 facts	Accuracy significantly degraded
KB Maximum	~500 facts	Unusable accuracy

Acronyms

Acronym	Full Name
GOS	Holographic Computation Engine
HDC	Hyperdimensional Computing
VSA	Vector Symbolic Architecture
KB	Knowledge Base
DSL	Domain-Specific Language
AST	Abstract Syntax Tree
LLM	Large Language Model

Related Pages