infotheory/aixi/
mod.rs

1//! AIXI Agent Implementations
2//!
3//! This module contains:
4//! - Monte Carlo AIXI (MC-AIXI) with pluggable predictive models.
5//! - AIQI (Universal AI with Q-Induction) with phase-indexed return prediction.
6//!
7//! AIXI is a theoretical mathematical formalism for universal artificial intelligence,
8//! which combines Solomonoff induction with sequential decision theory.
9//! This implementation follows the "Monte Carlo" approach (MC-AIXI) introduced by
10//! Veness et al., which uses Monte Carlo Tree Search (MCTS) to approximate
11//! the optimal policy.
12//!
13//! ## VM Backend
14//!
15//! A high-performance Firecracker-based VM environment is available via the `vm` feature:
16//!
17//! - **NyxVmEnvironment**: Uses nyx-lite for 10,000+ resets/second (requires KVM).
18
19pub mod agent;
20pub mod aiqi;
21pub mod common;
22pub mod environment;
23pub mod mcts;
24pub mod model;
25#[cfg(feature = "vm")]
26pub mod vm_nyx;