3 papers
cs.LG2026
Probabilistic Recurrent Intention Switching Model
Wenyuan Sheng, Hao Zhu, Joschka Boedecker
Inverse reinforcement learning (IRL) recovers reward functions from observed behavior, yet traditional methods assume a single stationary reward that cannot capture goal switching…
cs.CV2022
Pattern Attention Transformer with Doughnut Kernel
WenYuan Sheng
We present in this paper a new architecture, the Pattern Attention Transformer (PAT), that is composed of the new doughnut kernel. Compared with tokens in the NLP field, Transforme…
cs.CV2022
QKVA grid: Attention in Image Perspective and Stacked DETR
Wenyuan Sheng
We present a new model named Stacked-DETR(SDETR), which inherits the main ideas in canonical DETR. We improve DETR in two directions: simplifying the cost of training and introduci…