Open question · OPEN-REPRESENTATION-LOSS-ESTIMATION

Can representation and binning loss be estimated separately?

Openpublication: open

Meaning

Determine how to estimate both contributions rather than only their combined effect.

Statement

Construct cross-fitted estimators and uncertainty bars for I_R=Var(E[s|R]) and I_q=Var(E[s|q(R)]) when truth scores are available only on simulation.

Relationships

Related settled results

Representation loss and hard-quantization loss separateREPRESENTATION-QUANTIZATION-LOSS

The full chain beneath it

Where it comes from

Raised by

Representation loss and hard-quantization loss separateREPRESENTATION-QUANTIZATION-LOSS

Local graph

Prerequisites to the left, what it enables or raises to the right, boundary cases below. Open in the map.

The question in full

Estimating representation versus quantization loss

When truth scores exist only on simulation, devise cross-fitted estimators and uncertainty bars for $I_R=\operatorname{Var}(E[s\mid R])$ and $I_q=\operatorname{Var}(E[s\mid q(R)])$ .

From OPEN_PROBLEMS.md