DART: Distributional Adversarial Recurrent Training for Algorithm Learning
Distributional Adversarial Recurrent Training (DART), a training framework that replaces single-point supervision with a local target distribution around the ground-truth solution and aligns model outputs with this distribution through an adversarial objective, is proposed.
Hieu Tran Bao, Phung Thanh Dang, Minh Quang Nhat Pham et al.
· 0 citations