Scalable Multilabel Prediction via Randomized Methods
نویسندگان
چکیده
Modeling the dependence between outputs is a fundamental challenge in multilabel classification. In this work we show that a generic regularized nonlinearity mapping independent predictions to joint predictions is sufficient to achieve state-of-the-art performance on a variety of benchmark problems. Crucially, we compute the joint predictions without ever obtaining any independent predictions, while incorporating low-rank and smoothness regularization. We achieve this by leveraging randomized algorithms for matrix decomposition and kernel approximation. Furthermore, our techniques are applicable to the multiclass setting. We apply our method to a variety of multiclass and multilabel data sets, obtaining state-of-the-art results.
منابع مشابه
Comparing multilabel classification methods for provisional biopharmaceutics class prediction.
The biopharmaceutical classification system (BCS) is now well established and utilized for the development and biowaivers of immediate oral dosage forms. The prediction of BCS class can be carried out using multilabel classification. Unlike single label classification, multilabel classification methods predict more than one class label at the same time. This paper compares two multilabel method...
متن کاملFast Label Embeddings for Extremely Large Output Spaces
Many modern multiclass and multilabel problems are characterized by increasingly large output spaces. For these problems, label embeddings have been shown to be a useful primitive that can improve computational and statistical efficiency. In this work we utilize a correspondence between rank constrained estimation and low dimensional label embeddings that uncovers a fast label embedding algorit...
متن کاملEmbedding Inference for Structured Multilabel Prediction
A key bottleneck in structured output prediction is the need for inference during training and testing, usually requiring some form of dynamic programming. Rather than using approximate inference or tailoring a specialized inference method for a particular structure—standard responses to the scaling challenge— we propose to embed prediction constraints directly into the learned representation. ...
متن کاملHierarchical Multilabel Protein Function Prediction Using Local Neural Networks
Protein function predictions are usually treated as classification problems where each function is regarded as a class label. However, different from conventional classification problems, they have some specificities that make the classification task more complex. First, the problem classes (protein functions) are usually hierarchically structured, with superclasses and subclasses. Second, prot...
متن کاملFast Label Embeddings via Randomized Linear Algebra
Many modern multiclass and multilabel problems are characterized by increasingly large output spaces. For these problems, label embeddings have been shown to be a useful primitive that can improve computational and statistical efficiency. In this work we utilize a correspondence between rank constrained estimation and low dimensional label embeddings that uncovers a fast label embedding algorit...
متن کامل