What does it take to get *almost to the top? Meet Pei-Lien Chou, the worthy runner-up in our recent MLTrack Particle Tracking Challenge. We invited him to tell us about how he placed so well in this challenge.*

Out-of-Distribution Detection Using an Ensemble of Self Supervised Leave-out Classifiers

Yes, you can learn a lot from N=1, as long as you have some auxiliary information.

Google released Dataset Search to the world last week. Some asked for my thoughts on the new tool, and as you know, ask and you shall receive.

Mapillary is a collaborative street-level imagery platform that allows people and organizations to upload geo-tagged photos, which can then be used by customers to improve their mapping systems or applications.

Medical image processing is one of the key areas where deep learning is applied to great effect. Typical processing involves classification, detection, and segmentation using various medical image modalities. In this blog post, we outline a method to use the HIPAA Eligible service Amazon SageMaker to train a deep learning model for chest x-ray image classification using the Amazon SageMaker image classification algorithm. We demonstrate how this image classification algorithm can be an effective tool for analyzing high resolution medical images. We’ll use new features of the algorithm, such as multi-label support and mixed-precision training, to show how a chest x-ray image classification model can be trained 33 percent faster using mixed-precision mode compared to using float32 data type on Amazon EC2 P3 instances. We also show how chest x-ray images can be trained on high-resolution images, thus improving performance over low-resolution models

Shreeharsh Kelkar writes:

If buses arrive exactly every ten minutes, it’s true that your average wait time will be half that interval: 5 minutes. Qualitatively speaking, it’s easy to convince yourself that adding some variation to those arrivals will make the average wait time somewhat longer, as we’ll see here.

Frailty Models via Hierarchical Likelihood (frailtyHL)Implements the h-likelihood estimation procedures for general frailty models including competing-risk models and joint models.

In this paper, a taxonomy for memory networks is proposed based on their memory organization. The taxonomy includes all the popular memory networks: vanilla recurrent neural network (RNN), long short term memory (LSTM ), neural stack and neural Turing machine and their variants. The taxonomy puts all these networks under a single umbrella and shows their relative expressive power , i.e. vanilla RNN <=LSTM<=neural stack<=neural RAM. The differences and commonality between these networks are analyzed. These differences are also connected to the requirements of different tasks which can give the user instructions of how to choose or design an appropriate memory network for a specific task. As a conceptual simplified class of problems, four tasks of synthetic symbol sequences: counting, counting with interference, reversing and repeat counting are developed and tested to verify our arguments. And we use two natural language processing problems to discuss how this taxonomy helps choosing the appropriate neural memory networks for real world problem. A Taxonomy for Neural Memory Networks