# Self-Regulated Interactive Sequence-to-Sequence Learning

## TL;DR

This paper introduces a self-regulation approach for sequence-to-sequence learning that optimizes feedback strategies to improve cost-effectiveness and robustness in neural machine translation.

## Contribution

It formulates self-regulated feedback selection as a learning-to-learn problem, enabling adaptive, cost-aware sequence-to-sequence training with demonstrated advantages over traditional methods.

## Key findings

- Self-regulator discovers an $psilon$-greedy strategy for feedback.
- Improves cost-quality trade-off in neural machine translation.
- Shows robustness under domain shift.

## Abstract

Not all types of supervision signals are created equal: Different types of feedback have different costs and effects on learning. We show how self-regulation strategies that decide when to ask for which kind of feedback from a teacher (or from oneself) can be cast as a learning-to-learn problem leading to improved cost-aware sequence-to-sequence learning. In experiments on interactive neural machine translation, we find that the self-regulator discovers an $\epsilon$-greedy strategy for the optimal cost-quality trade-off by mixing different feedback types including corrections, error markups, and self-supervision. Furthermore, we demonstrate its robustness under domain shift and identify it as a promising alternative to active learning.