Combinets: Learning New Classifiers via Recombination

Problems with few examples of a new class of objects prove challenging to most classifiers. One solution to is to reuse existing data through transfer methods such as one-shot learning or domain adaption. However these approaches require an explicit hand-authored or learned definition of how reuse can occur. We present an approach called conceptual expansion that learns how to reuse existing machine-learned knowledge when classifying new cases. We evaluate our approach by adding new classes of objects to the CIFAR-10 dataset and varying the number of available examples of these new classes.

[1]  Li Fei-Fei,et al.  ImageNet: A large-scale hierarchical image database , 2009, CVPR.

[2]  Ieee Xplore,et al.  IEEE Transactions on Pattern Analysis and Machine Intelligence Information for Authors , 2022, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[3]  Santiago Ontañón,et al.  Amalgam-Based Reuse for Multiagent Case-Based Reasoning , 2011, ICCBR.

[4]  Ashok K. Goel,et al.  Meta-case-Based Reasoning: Using Functional Models to Adapt Case-Based Agents , 2001, ICCBR.

[5]  Rüdiger Kapitza,et al.  Proceedings of the 3rd International DiscCoTec Workshop on Middleware-Application Interaction , 2009 .

[6]  Christoph H. Lampert,et al.  Learning to detect unseen object classes by between-class attribute transfer , 2009, 2009 IEEE Conference on Computer Vision and Pattern Recognition.

[7]  Trevor Darrell,et al.  What you saw is not what you get: Domain adaptation using asymmetric kernel transforms , 2011, CVPR 2011.

[8]  Boyang Li,et al.  Goal-Driven Conceptual Blending: A Computational Approach for Creativity , 2012, ICCC.

[9]  Hal Daumé,et al.  Frustratingly Easy Domain Adaptation , 2007, ACL.

[10]  Matthew Guzdial,et al.  Learning to Blend Computer Game Levels , 2016, ICCC.

[11]  Yifan Gong,et al.  Large-Scale Domain Adaptation via Teacher-Student Learning , 2017, INTERSPEECH.

[12]  Peter Lee.  The Unreasonable Effectiveness, and Difficulty, of Data in Healthcare , 2019, KDD.

[13]  The Bad and the Ugly? , 2005, Science.

[14]  Weiwei Zhang,et al.  Cat Head Detection - How to Effectively Exploit Shape and Texture Features , 2008, ECCV.

[15]  Mark J. F. Gales,et al.  Sequence Student-Teacher Training of Deep Neural Networks , 2016, INTERSPEECH.

[16]  Samy Bengio,et al.  Zero-Shot Learning by Convex Combination of Semantic Embeddings , 2013, ICLR.

[17]  Wojciech Zaremba,et al.  Improved Techniques for Training GANs , 2016, NIPS.

[18]  Terri Gullickson.  The Creative Mind: Myths and Mechanisms. , 1995 .

[19]  Jean Lieber,et al.  Belief Merging-Based Case Combination , 2009, ICCBR.

[20]  Sébastien Konieczny,et al.  Logic Based Merging , 2011, J. Philos. Log..

[21]  Ilja Kuzborskij,et al.  Stability and Hypothesis Transfer Learning , 2013, ICML.

[22]  Sanja Fidler,et al.  Predicting Deep Zero-Shot Convolutional Neural Networks Using Textual Descriptions , 2015, 2015 IEEE International Conference on Computer Vision (ICCV).

[23]  Pietro Perona,et al.  The Caltech-UCSD Birds-200-2011 Dataset , 2011 .

[24]  Soumith Chintala,et al.  Unsupervised Representation Learning with Deep Convolutional Generative Adversarial Networks , 2015, ICLR.

[25]  Ahmed M. Elgammal,et al.  Link the Head to the "Beak": Zero Shot Learning from Noisy Text Description at Part Precision , 2017, 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[26]  Jeffrey M. Bradshaw,et al.  Ieee Intelligent Systems Kinds of Systems? , 2009 .

[27]  Penousal Machado,et al.  A Pig, an Angel and a Cactus Walk Into a Blender: A Descriptive Approach to Visual Blending , 2017, ICCC.

[28]  Agnar Aamodt,et al.  Evidence-Driven Retrieval in Textual CBR: Bridging the Gap Between Retrieval and Reuse , 2015, ICCBR.

[29]  Bernt Schiele,et al.  Zero-Shot Learning — The Good, the Bad and the Ugly , 2017, 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[30]  François Laviolette,et al.  Domain-Adversarial Training of Neural Networks , 2015, J. Mach. Learn. Res..

[31]  Ilja Kuzborskij,et al.  From N to N+1: Multiclass Transfer Incremental Learning , 2013, 2013 IEEE Conference on Computer Vision and Pattern Recognition.

[32]  Subhransu Maji,et al.  Fine-Grained Visual Classification of Aircraft , 2013, ArXiv.

[33]  Omer Levy,et al.  Teaching Machines to Learn by Metaphors , 2012, AAAI.

[34]  D. Gentner,et al.  The analogical mind : perspectives from cognitive science , 2001 .

[35]  Martial Hebert,et al.  Learning to Learn: Model Regression Networks for Easy Small Sample Learning , 2016, ECCV.

[36]  Ali Farhadi,et al.  Learning Everything about Anything: Webly-Supervised Visual Concept Learning , 2014, 2014 IEEE Conference on Computer Vision and Pattern Recognition.

[37]  Paul Thagard,et al.  The AHA! Experience: Creativity Through Emergent Binding in Neural Networks , 2011, Cogn. Sci..

[38]  Santiago Ontañón,et al.  Amalgams: A Formal Approach for Combining Multiple Case Solutions , 2010, ICCBR.

[39]  Raquel Hervás,et al.  Case-Based Reasoning for Knowledge-Intensive Template Selection During Text Generation , 2006, ECCBR.

[40]  Alex Krizhevsky,et al.  Learning Multiple Layers of Features from Tiny Images , 2009 .

[41]  Zachary Chase Lipton,et al.  Born Again Neural Networks , 2018, ICML.

[42]  R. A. M. O N L O P E Z D E M A N T A R A S,et al.  Retrieval, reuse, revision and retention in case-based reasoning , 2006 .

[43]  Wei-Lun Chao,et al.  An Empirical Study and Analysis of Generalized Zero-Shot Learning for Object Recognition in the Wild , 2016, ECCV.

[44]  Jorge Fox,et al.  Exploring approaches to dynamic adaptation , 2009, MAI '09.

[45]  Ralph Bergmann,et al.  Techniques and Knowledge Used for Adaptation During Case-Based Problem Solving , 1998, IEA/AIE.