Please use this identifier to cite or link to this item: https://hdl.handle.net/2440/117897
Type: Conference paper
Title: Joint learning of set cardinality and state distribution
Author: Rezatofighi, H.
Milan, A.
Shi, Q.
Dick, A.
Reid, I.
Citation: Proceedings of the ... AAAI Conference on Artificial Intelligence. AAAI Conference on Artificial Intelligence, 2018, pp.3968-3975
Publisher: AAAI
Publisher Place: online
Issue Date: 2018
Series/Report no.: AAAI Conference on Artificial Intelligence
ISBN: 9781577358008
ISSN: 2159-5399
2374-3468
Conference Name: AAAI Conference on Artificial Intelligence (AAAI) (2 Feb 2018 - 7 Feb 2018 : New Orleans, USA)
Statement of
Responsibility: 
S. Hamid Rezatofighi, Anton Milan, Qinfeng Shi, Anthony Dick, Ian Reid
Abstract: We present a novel approach for learning to predict sets using deep learning. In recent years, deep neural networks have shown remarkable results in computer vision, natural language processing and other related problems. Despite their success, traditional architectures suffer from a serious limitation in that they are built to deal with structured input and output data, i.e. vectors or matrices. Many real-world problems, however, are naturally described as sets, rather than vectors. Existing techniques that allow for sequential data, such as recurrent neural networks, typically heavily depend on the input and output order and do not guarantee a valid solution. Here, we derive in a principled way, a mathematical formulation for set prediction where the output is permutation invariant. In particular, our approach jointly learns both the cardinality and the state distribution of the target set. We demonstrate the validity of our method on the task of multi-label image classification and achieve a new state of the art on the PASCAL VOC and MS COCO datasets.
Rights: Copyright © 2018, Association for the Advancement of Artificial Intelligence (www.aaai.org). All rights reserved.
Appears in Collections:Aurora harvest 8
Australian Institute for Machine Learning publications
Computer Science publications

Files in This Item:
There are no files associated with this item.


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.