Order Preserving Sparse Coding | IEEE Journals & Magazine | IEEE Xplore

Sparse coding (SC) has been successfully used in various computer vision applications [1], [2], [3]. Using sparse coding for classification tasks presents two advantages: 1) the method is training-free, if the dictionary is constructed by simply sampling from the training data; and 2) the method provides flexibility in handling dynamic training settings, i.e., new training samples and new class labels. Sparse coding compactly represents an objects as a linear combination of a small number of elements in a dictionary. Data are usually structured. For example, there may exist group structures in the data. To handle such structures, two alternative methods, Elastic Net [4] and group Lasso [5] have been proposed. They favor the selection of a small number of groups of correlated dictionary samples to represent the testing data. Another example is when data are described by several types of features (modalities). In object recognition, we may extract K types of image representations from

Order Preserving Sparse Coding | IEEE Journals & Magazine | IEEE Xplore --> --> IEEE Account Change Username/Password Update Address Purchase Details Payment Options Order History View Purchased Documents Profile Information Communications Preferences Profession and Education Technical Interests Need Help? US & Canada: +1 800 678 4333 Worldwide: +1 732 981 0060 Contact & Support About IEEE Xplore Contact Us Help Accessibility Terms of Use Nondiscrimination Policy Sitemap Privacy & Opting Out of Cookies A not-for-profit organization, IEEE is the world's largest technical professional organizati

related reading

• Unsupervised Feature Learning and Deep Learning Tutorialufldl.stanford.edu
• machine learning - Sparse coding vs. sparse PCA, are they the same thing? - Cross Validatedstats.stackexchange.com
• Towards Monosemanticity: Decomposing Language Models With Dictionary Learningtransformer-circuits.pub
• Compressed sensing - Wikipediaen.wikipedia.org
• Scaling Monosemanticity: Extracting Interpretable Features from Claude 3 Sonnettransformer-circuits.pub
• Sparse Autoencoders Find Highly Interpretable Features in Language Modelsarxiv.org
• [Interim research report] Taking features out of superposition with sparse autoencoders — LessWronglesswrong.com
• Sparse PCA - Wikipediaen.wikipedia.org
• [2410.12101] The Persian Rug: solving toy models of superposition using large-scale symmetriesarxiv.org
• Sparse Crosscoders for Cross-Layer Features and Model Diffingtransformer-circuits.pub
• Identifying Functionally Important Features with End-to-End Sparse Dictionary Learning — LessWronglesswrong.com
• Bag-of-words model - Wikipediaen.wikipedia.org