Vinod Anbalagan

The Lab · 0 of 6 toward the Notch paper

Teaching a world model
what to remember.

Toward one question: what should a robot remember? And what should decide what goes into a small memory, without giving up reliability or speed?

01

One question

Isolate a single mechanism and ask when it appears or breaks.

02

A scaling ladder

Three or four points along one axis: does the effect hold, grow, or vanish?

03

Honest limits

Every write-up ends with what would still need testing at scale.

04

Runs in your browser

Trained in Python, exported, and run live on this page. No server, no waiting.

The Notch paper

  1. 01
    FoundationsIn progress

    The Baseline

    Reproduce LeWorldModel, a small JEPA world model that plans from pixels, on one of its tasks. Measure success rate and planning time per decision on my own hardware, and check whether the planner can show the plan it held before and after each new observation.

    Scaling ladder · planning horizon 5 → 10 → 20 steps

  2. 03
    SignalsPlanned

    The Signal Audit

    Before building any memory: log two signals, the world model's surprise and whether each observation changed the plan, and compare them with labelled moments that matter, including under distractors, recolouring and repeated frames. Which signal marks what is worth remembering?

    Scaling ladder · change that matters → distractor → appearance only → own progress

  3. 04
    MemoryPlanned

    Remembering Everything

    The expensive answers: stack recent frames, give the predictor a recurrent state, add always-on learned slots, or keep every past state and attend over all of it, plus an oracle handed the hidden state. These set the ceiling and the price.

    Scaling ladder · history length 4 → 16 → full episode

  4. 05
    SelectionPlanned

    The Write Rule

    One model, one memory budget, five rules for what gets written: time, chance, surprise, plan change, and surprise weighted by plan change. Which fills a small memory best, and does what it keeps really guide the plan?

    Scaling ladder · time → chance → surprise → plan change → both

  5. 06
    SpeedPlanned

    The Price of Memory

    Success rate against planning time per decision as episodes get longer, for every memory variant, over several seeds. Where does each kind of memory sit on the trade-off between reliability and speed?

    Scaling ladder · episode length short → long → very long

  6. →
    CapstoneResearch

    Notch

    What should a robot remember? Comparing what should decide what a world-model planner writes into a small, fixed memory, measured on both success and speed.

    The write-up of all 6

After the paper