infotheory/aixi/
mod.rs

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