Docs  /  MathNN Learning Functions (full stack)

MathNN Learning Functions (full stack)

Implementation of the learning-functions brainstorm (L0–L5).

Architecture

nn learn <verb>
    → _MathNNLearning
         ├─ _MathNNExperience  (transitions, demos, prefer pairs, jobs)
         ├─ _RLAgentNet        (Q/RM/WM, Select/Observe, SupervisedQStep)
         ├─ _MathNNDevice      (train/infer placement)
         └─ _RLCheckpointMySQL (meta, transfer, quant, federate, promote)

Experience store

Buffer Filled by
Transitions Every RLAgentNet::Observe
Last state/action Select + Observe
Demos imitate start + Select, or load file
Prefer pairs prefer good/bad and prefer pair
Jobs nn learn jobs enqueue …

Verbs (implemented)

Verb Real behavior
prefer RM feedback on last state; auto pair buffer
prefer pair / train Explicit pairs + batch RM train
curious WM on + scale; warm on stored transitions
imitate start/stop/train Demo record + BC via SupervisedQStep
distill Weight copy + KD matching teacher Q
transfer Full load; optional encoder fine-tune
adapt RM plasticity + replay mix
quantize Save + cal pass + NPU placement + JSON sidecar
shadow Peer agent + dual-Q metrics on Select
curriculum Phases + success-gated next
hierarchy Options + HierarchyAct worker steps
whatif Q on last state + WM-style horizon
align RM + low curiosity + batch caution labels
federate push/pull MySQL + avg ≈ KD toward global
promote multi-metric gate (loss, pref, shadow)
scale tiny/default/large/huge + device pref
pack TutorOS, EmbodiedCraft, LifeCoach, Operator, MetaMind, WorldForge
jobs enqueue / drain simple train jobs
experience Store stats

Examples

nn learn imitate start life
# … agent acts …
nn learn imitate stop life
nn learn imitate train life 128

nn learn prefer good
nn learn prefer train life 32

nn learn distill life from life_focus_blocks 64
nn learn transfer minecraft space_engineers encoder
nn learn shadow minecraft on
nn learn promote minecraft multi
nn learn pack EmbodiedCraft
nn learn pack LifeCoach
nn learn whatif 0 3
nn learn federate life push
nn learn federate life avg

Files

  • MathNNExperience.hpp/cpp
  • MathNNLearning.hpp/cpp
  • hooks in RLAgentNet.cpp (Observe → store, Select → demo/shadow)

Extended methods (NeuralLearnMethods)

See NeuralLearnMethods.md:

nn learn method pack learn_all
nn learn method dyna on 8
nn learn method offline 128
nn learn method nightly minecraft 256

Dyna imagination, offline RL, PER, n-step, DAgger, EWC-lite, SSL mask, RLAIF labels, multi-objective, nightly jobs.

Still future (honest)

  • True ONNX NPU EP execution
  • Full sum-tree PER and true PPO clip
  • External LoRA training job (export is ready; train offline)
  • Cryptographic FedAvg / secure aggregation
  • Full DAgger loop UI
  • External async trainer process
Generated from the project markdown docs on 2026-07-24. This is a static, self-contained site.