International Conference on Data Mining (ICDM) google
The IEEE International Conference on Data Mining series (ICDM) has established itself as the world’s premier research conference in data mining. It provides an international forum for presentation of original research results, as well as exchange and dissemination of innovative, practical development experiences. The conference covers all aspects of data mining, including algorithms, software and systems, and applications. ICDM draws researchers and application developers from a wide range of data mining related areas such as statistics, machine learning, pattern recognition, databases and data warehousing, data visualization, knowledge-based systems, and high performance computing. By promoting novel, high quality research findings, and innovative solutions to challenging data mining problems, the conference seeks to continuously advance the state-of-the-art in data mining. Besides the technical program, the conference features workshops, tutorials, panels and, since 2007, the ICDM data mining contest. …

Mutual Information Neural Estimator (MINE) google
We argue that the estimation of the mutual information between high dimensional continuous random variables is achievable by gradient descent over neural networks. This paper presents a Mutual Information Neural Estimator (MINE) that is linearly scalable in dimensionality as well as in sample size. MINE is back-propable and we prove that it is strongly consistent. We illustrate a handful of applications in which MINE is succesfully applied to enhance the property of generative models in both unsupervised and supervised settings. We apply our framework to estimate the information bottleneck, and apply it in tasks related to supervised classification problems. Our results demonstrate substantial added flexibility and improvement in these settings. …

StrassenNets google
A large fraction of the arithmetic operations required to evaluate deep neural networks (DNNs) are due to matrix multiplications, both in convolutional and fully connected layers. Matrix multiplications can be cast as $2$-layer sum-product networks (SPNs) (arithmetic circuits), disentangling multiplications and additions. We leverage this observation for end-to-end learning of low-cost (in terms of multiplications) approximations of linear operations in DNN layers. Specifically, we propose to replace matrix multiplication operations by SPNs, with widths corresponding to the budget of multiplications we want to allocate to each layer, and learning the edges of the SPNs from data. Experiments on CIFAR-10 and ImageNet show that this method applied to ResNet yields significantly higher accuracy than existing methods for a given multiplication budget, or leads to the same or higher accuracy compared to existing methods while using significantly fewer multiplications. Furthermore, our approach allows fine-grained control of the tradeoff between arithmetic complexity and accuracy of DNN models. Finally, we demonstrate that the proposed framework is able to rediscover Strassen’s matrix multiplication algorithm, i.e., it can learn to multiply $2 \times 2$ matrices using only $7$ multiplications instead of $8$. …

Advertisements