Coactive learning for interactive machine translation
Artem Sokolov, Stefan Riezler, Shay B. Cohen · 2015
Coactive learning describes the interaction be-tween an online structured learner and a human user who corrects the learner by responding with weak feedback, that is, with an improved, but not necessarily optimal, structure. We apply this framework to discriminative learning in interac-tive machine translation. We present a gener-alization to latent variable models and give re-gret and generalization bounds for online learn-ing with a feedback-based latent perceptron. We show experimentally that learning from weak feedback in machine translation leads to conver-gence in regret and translation error. 1