Self-Paced Curriculum Learning

Curriculum learning (CL) or self-paced learning (SPL) represents a recently proposed learning regime inspired by the learning process of humans and animals that gradually proceeds from easy to more complex samples in training. The two methods share a similar conceptual learning paradigm, but differ in specific learning schemes. In CL, the curriculum is predetermined by prior knowledge, and remain fixed thereafter. Therefore, this type of method heavily relies on the quality of prior knowledge while ignoring feedback about the learner. In SPL, the curriculum is dynamically determined to adjust to the learning pace of the leaner. However, SPL is unable to deal with prior knowledge, rendering it prone to overfitting. In this paper, we discover the missing link between CL and SPL, and propose a unified framework named self-paced curriculum leaning (SPCL). SPCL is formulated as a concise optimization problem that takes into account both prior knowledge known before training and the learning progress during training. In comparison to human education, SPCL is analogous to "instructor-student-collaborative" learning mode, as opposed to "instructor-driven" in CL or "student-driven" in SPL. Empirically, we show that the advantage of SPCL on two tasks.

[1]  Hideki Hayakawa.  Photometric stereo under a light source with arbitrary motion , 1994 .

[2]  Jorge Nocedal,et al.  Algorithm 778: L-BFGS-B: Fortran subroutines for large-scale bound-constrained optimization , 1997, TOMS.

[3]  Takeo Kanade,et al.  Shape and motion from image streams under orthography: a factorization method , 1992, International Journal of Computer Vision.

[4]  Kathrin Klamroth,et al.  Biconvex sets and optimization with biconvex functions: a survey and extensions , 2007, Math. Methods Oper. Res..

[5]  Jason Weston,et al.  Curriculum learning , 2009, ICML '09.

[6]  Valentin I. Spitkovsky,et al.  Baby Steps: How “Less is More” in Unsupervised Dependency Parsing , 2009 .

[7]  Daphne Koller,et al.  Self-Paced Learning for Latent Variable Models , 2010, NIPS.

[8]  Daphne Koller,et al.  Learning specific-class segmentation from diverse data , 2011, 2011 International Conference on Computer Vision.

[9]  A. Smeaton,et al.  TRECVID 2013 -- An Overview of the Goals, Tasks, Data, Evaluation Mechanisms, and Metrics | NIST , 2011 .

[10]  Bilge Mutlu,et al.  How Do Humans Teach: On Curriculum Learning and Teaching Dimension , 2011, NIPS.

[11]  Alexander G. Hauptmann,et al.  Leveraging high-level and low-level features for multimedia event detection , 2012, ACM Multimedia.

[12]  Yang Gao,et al.  Self-paced dictionary learning for image classification , 2012, ACM Multimedia.

[13]  Jingdong Wang,et al.  A Probabilistic Approach to Robust Matrix Factorization , 2012, ECCV.

[14]  Fei-Fei Li,et al.  Shifting Weights: Adapting Object Detectors from Image to Video , 2012, NIPS.

[15]  Lei Zhang,et al.  A Cyclic Weighted Median Method for L1 Low-Rank Matrix Factorization with Missing Entries , 2013, AAAI.

[16]  Alexandre Bernardino,et al.  Unifying Nuclear Norm and Bilinear Factorization Approaches for Low-Rank Matrix Decomposition , 2013, 2013 IEEE International Conference on Computer Vision.

[17]  Pascal Vincent,et al.  Representation Learning: A Review and New Perspectives , 2012, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[18]  Deva Ramanan,et al.  Self-Paced Learning for Long-Term Tracking , 2013, 2013 IEEE Conference on Computer Vision and Pattern Recognition.

[19]  Hui Cheng,et al.  Video event recognition using concept attributes , 2013, 2013 IEEE Workshop on Applications of Computer Vision (WACV).

[20]  Alexander I. Rudnicky,et al.  Using conversational word bursts in spoken term detection , 2013, INTERSPEECH.

[21]  Yi Yang,et al.  E-LAMP: integration of innovative ideas for multimedia event detection , 2013, Machine Vision and Applications.

[22]  John R. Kender,et al.  Learning by focusing: A new framework for concept recognition and feature selection , 2013, 2013 IEEE International Conference on Multimedia and Expo (ICME).

[23]  Sumit Basu,et al.  Teaching Classification Boundaries to Humans , 2013, AAAI.

[24]  Florian Metze,et al.  Deep maxout networks for low-resource speech recognition , 2013, 2013 IEEE Workshop on Automatic Speech Recognition and Understanding.

[25]  Florian Metze,et al.  Improvements to speaker adaptive training of deep neural networks , 2014, 2014 IEEE Spoken Language Technology Workshop (SLT).

[26]  Shiguang Shan,et al.  Self-Paced Learning with Diversity , 2014, NIPS.

[27]  Yoshua Bengio,et al.  Evolving Culture Versus Local Minima , 2014, Growing Adaptive Machines.

[28]  Koichi Shinoda,et al.  n-gram Models for Video Semantic Indexing , 2014, ACM Multimedia.

[29]  Lei Zhang,et al.  Robust Principal Component Analysis with Complex Noise , 2014, ICML.

[30]  Benoit Huet,et al.  When textual and visual information join forces for multimedia retrieval , 2014, ICMR.

[31]  Vasileios Mezaris,et al.  Video event detection using generalized subclass discriminant analysis and linear support vector machines , 2014, ICMR.

[32]  Deyu Meng,et al.  Easy Samples First: Self-paced Reranking for Zero-Example Multimedia Search , 2014, ACM Multimedia.

[33]  Chong-Wah Ngo,et al.  Video Event Detection Using Motion Relativity and Feature Selection , 2014, IEEE Transactions on Multimedia.

[34]  Alexander G. Hauptmann,et al.  Instructional Videos for Unsupervised Harvesting and Learning of Action Examples , 2014, ACM Multimedia.

[35]  Yang Yang,et al.  Start from Scratch: Towards Automatically Identifying, Modeling, and Naming Visual Attributes , 2014, ACM Multimedia.

[36]  Teruko Mitamura,et al.  Zero-Example Event Search using MultiModal Pseudo Relevance Feedback , 2014, ICMR.

[37]  Song Tan,et al.  Placing Videos on a Semantic Hierarchy for Search Result Navigation , 2014, TOMM.

[38]  Sotirios Chatzis.  Dynamic Bayesian Probabilistic Matrix Factorization , 2014, AAAI.