Personalized handwriting recognition via biased regularization

By Wolf Kienzle, Kumar Chellapilla. 2006

71 citations recorded by the source corpus.

Subfields: Computer Vision and Pattern Recognition, Signal Processing

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computer Vision and Pattern Recognition

In Signal Processing

Citations in the corpus

Overcoming Catastrophic Forgetting by Incremental Moment Matching (2017, methodology): “ kis added to the following objective function for ﬁnding k, where is a hyperparameter: logp(y kjX k; k) jj k k 1jj 2 2 (10) The concept of L2-transfer is commonly used in transfer learning [25, 26] and continual learning [7, 8] with large . Unlike the previous usage of large , we use small for t”
Continual Learning by Using Information of Each Class Holistically (2021, background): “1: i (cid:0) 1 , to initialize (cid:18) i , and L2-transfer in (Evgeniou and Pontil 2004; Kienzle and Chellapilla 2006) is also added to Eq.”
A Convex Formulation for Learning Task Relationships in Multi-Task Learning (2010, background): “Most of the above methods focus on symmetric multi-task learning, but there also exist some previous works that study asymmetric multi-task learning [23, 19, 10].”
Transfer Learning in Collaborative Filtering for Sparsity Reduction (2010, unknown): “CST can also be considered as a two-sided extension in matrix form of onesided domain adaptation methods in vector form (Kienzle and Chellapilla 2006), which is proposed not for collaborative filtering but classification problems and achieve knowledge transfer via incorporating the model parameters…”
Learning to Learn: Model Regression Networks for Easy Small Sample Learning (2016, methodology): “In addition, adaptive SVM and its variants present SVM-based model adaptation by combining classifiers learned on related categories [47, 102, 23, 2, 3, 53, 88, 97].”
A Theoretical Analysis of Metric Hypothesis Transfer Learning (2015, methodology): “For example in (Kienzle & Chellapilla, 2006), the authors propose to use biased regular- A Theoretical Analysis of Metric Hypothesis Transfer Learning ization to learn SVM classifiers.”
Gesture Recognition Portfolios for Personalization (2014, methodology): “This type of user adaptation is used extensively in speech [12, 17] and handwriting recognition [3, 11], with the obvious tradeoff between the amount of personalization data (and subsequent accuracy) versus the inconvenience to the user.”
BNS: Building Network Structures Dynamically for Continual Learning (2021, background): “When the action is "reuse" or "fuse", in order to ensure that these parameters will not change too much and thus alleviate forgetting when training the current task, we introduce L2 transfer regularization [12, 24] to constrain the training of the corresponding layer’s parameters.”
Attend to You: Personalized Image Captioning with Context Sequence Memory Networks (2017, background): “There have been many studies about personalization in computer vision and natural language processing [2, 8, 1, 32, 10, 23].”