PICASO: Permutation-Invariant Cascaded Attentional Set Operator
This addresses a limitation in set-input networks for tasks like clustering and anomaly detection, offering a novel module with broad applicability, though it is incremental as it builds on existing attention-based methods.
The paper tackled the problem of set-input deep networks being ineffective against common data variations like image translation or viewpoint change, and proposed PICASO, a permutation-invariant cascaded attentional set operator, which improved SmallNORB classification accuracy by about 10% points, CelebA anomaly detection ROC/PR by 22%/10%, and CLEVR state prediction AP by about 40%.
Set-input deep networks have recently drawn much interest in computer vision and machine learning. This is in part due to the increasing number of important tasks such as meta-learning, clustering, and anomaly detection that are defined on set inputs. These networks must take an arbitrary number of input samples and produce the output invariant to the input set permutation. Several algorithms have been recently developed to address this urgent need. Our paper analyzes these algorithms using both synthetic and real-world datasets, and shows that they are not effective in dealing with common data variations such as image translation or viewpoint change. To address this limitation, we propose a permutation-invariant cascaded attentional set operator (PICASO). The gist of PICASO is a cascade of multihead attention blocks with dynamic templates. The proposed operator is a stand-alone module that can be adapted and extended to serve different machine learning tasks. We demonstrate the utilities of PICASO in four diverse scenarios: (i) clustering, (ii) image classification under novel viewpoints, (iii) image anomaly detection, and (iv) state prediction. PICASO increases the SmallNORB image classification accuracy with novel viewpoints by about 10% points. For set anomaly detection on CelebA dataset, our model improves the areas under ROC and PR curves dataset by about 22% and 10%, respectively. For the state prediction on CLEVR dataset, it improves the AP by about 40%.