DataCupid: Stable Matching Repair for Training Data Acquisition with Delayed Feedback

Key points

• We study announced invalidations with bounded, possibly reward-dependent reporting delays and pending evaluations that do not occupy future matching capacity.1
• Residualload repair (RLR) combines sound outcome certificates with batched matching schedules for unequal sample deficits.1
• On fixed priority-star markets with unknown Bernoulli gaps, PrioritySearch attains the history-adjusted first-order constant for expected certification time as δ {\sim} 0, under hypothesis-independent linear history scaling and sublogarithmic reporting delays.1
• Paired simulations and measured-feedback replay quantify the trade-offs: in the replay, RLR-P has lower mean capped update-certification time than a monitored width-based baseline, at higher oracle CPU cost.1

Sources (1)

• [1]DataCupid: Stable Matching Repair for Training Data Acquisition with Delayed Feedback
  arXiv (AI, ML, NLP, CV, robotics, multi-agent) · Oct 9, 03:59 AM

  Matching repair after utility changes can reuse valid observations, but new evaluations must respect matching capacity and may return out of order.

  We study announced invalidations with bounded, possibly reward-dependent reporting delays and pending evaluations that do not occupy future matching capacity.

Extractive summary: sentences quoted from the sources.