Brief Introduction

Self-paced learning (SPL) is a recently proposed learning regime inspired by the learning process of humans and animals that gradually incorporates easy to more complex samples into training. While several easy SPL implementation strategies have been proposed, it is still short of a general paradigm for guiding the construction of rational SPL learning regimes targeting specific applications. To resolve this problem, we provide an axiom for insightfully formulating the underlying principles of self-paced learning. This axiomatic understanding not only involves the previous SPL learning schemes as its special cases, but also can be utilized to extend a series of new SPL implementation regimes based on certain application aims. In the recent two years, we have constructed several SPL realizations, including SPaR, SPLD, SPCL, SPMF, SPMIL, based on this axiom, and achieved the best performance in several known benchmark datasets, e.g., Web Query, Hollywood2, and Olympic Sports. Especially, this paradigm has been integrated into the system developed by CMU Informedia team, and achieved the leading performance in challenging semantic query (SQ)/000Ex tasks of the TRECVID MED/MER competition organized by NIST in 2014.

Related Publications

This paper provides some fundamental understanding of various SPL regimes.

Deyu Meng, Qian Zhao, Lu Jiang. A Theoretical Understanding of Self-paced Learning. accepted by Information Sciences, 2017 [Arxiv version] [Slides].

 

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

More SPL realizations:

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

ASPL:

Liang Lin, Keze Wang, Deyu Meng, Wangmeng Zuo, Lei Zhang. Active Self-Paced Learning for Cost-Effective and Progressive Face Identification. To appear in IEEE Transactions on Pattern Analysis and Machine Intelligence, 2017. [arxiv version]

 

Three SPL papers in IJCAI 2016:

[1] Junwei Liang, Lu Jiang, Deyu Meng and Alex Hauptmann, Learning to Detect Concepts from Webly-Labeled Video Data. To appear in International Joint Conference on Artificial Intelligence (IJCAI), 2016.

[2] Dingwen Zhang, Deyu Meng, Long Zhao and Junwei Han, Bridging Saliency Detection to Weakly Supervised Object Detection Based on Self-paced Curriculum Learning. To appear in International Joint Conference on Artificial Intelligence (IJCAI), 2016.

[3] Te Pi, Xi Li, Zhongfei Zhang, Deyu Meng, Fei Wu, Jun Xiao and Yueting Zhuang, Self-Paced Boost Learning for Classification. To appear in International Joint Conference on Artificial Intelligence (IJCAI), 2016.

 

MOSPL:

Hao Li, Maoguo Gong, Deyu Meng, Qiguang Miao. Multi-optimization Self-paced Learning. AAAI, 2015.

 

SPaR:

Lu Jiang, Deyu Meng, Teruko Mitamura, Alexander Hauptmann. Easy Samples First: Self-paced Reranking for Zero-Example Multimedia Search. ACM MM. 2014. Slides.

  

SPMF:

Qian Zhao, Deyu Meng, Lu Jiang, Qi Xie, Zongben Xu, Alexander Hauptmann. Self-paced Matrix Factorization. AAAI, 2015. Supplementary material.

 

SPCL:

Lu Jiang, Deyu Meng, Qian Zhao, Shiguang Shan, Alexander Hauptmann. Self-paced Curriculum Learning. AAAI, 2015 (oral).Supplementary material, slides, code.

  

SPMIL:

D. Zhang, D. Meng, C. Li, L. Jiang, Q. Zhao, and J. Han. A Self-paced Multiple-instance Learning Framework for Co-saliency Detection. In ICCV, 2015.

Dingwen Zhang, Deyu Meng, Junwei Han. Co-saliency Detection via A Self-paced Multiple-instance Learning Framework. To appear in IEEE Transactions on Pattern Analysis and Machine Intelligence, 2016.

 

SPLD:

Lu Jiang, Deyu Meng, Shoou-I Yu, Zhen-Zhong Lan, Shiguang Shan, Alexander Hauptmann.Self-paced Learning with Diversity. NIPS, 2014.Supplementary material, code.

 

SPL has been integrated into the system developed by CMU Informedia team, and achieved the leading performance in challenging semantic query (SQ)/000Ex tasks of the TRECVID MED/MER competition organized by NIST in 2014:

CMU-Informedia@ TRECVID 2014. Shoou-I Yu, Lu Jiang, Zhongwen Xu, Zhenzhong Lan, Shicheng Xu, Xiaojun Chang, Xuanchong Li, Zexi Mao, Chuang Gan, Yajie Miao, Xingzhong Du, Yang Cai, Lara Martin, Nikolas Wolfe, Anurag Kumar, Huan Li, Ming Lin, Zhigang Ma, Yi Yang, Deyu Meng, Shiguang Shan, Pinar Duygulu Sahin, Susanne Burger, Florian Metze, Rita Singh, Bhiksha Raj, Teruko Mitamura, Richard Stern and Alexander Hauptmann. In TRECVID Video Retrieval Evaluation Workshop, NIST, 2014.

  

The paper introducing the Informedia Ex0 system achieved the best-paper-runner-up in ICMR 2015:

Lu Jiang, Shoou-I Yu, Deyu Meng, Teruko Mitamura, Alexander Hauptmann. Bridging the Ultimate Semantic Gap: A Semantic Search Engine for Internet Videos. In ACM International Conference on Multimedia Retrieval (ICMR). 2015. [BibTex |supplementary materials | slides | project page] Best paper runner-up 
[featured in]
Pittsburgh Supercomputing Center