Neural Architecture Search (NAS) aims at finding one "single" architecture that achieves the best accuracy for a given task such as image recognition.In this paper, we study the instance-level variation,and demonstrate that instance-awareness is an important yet currently missing component of NAS. Based on this observation, we propose InstaNAS for searching toward instance-level architectures;the controller is trained to search and form a "distribution of architectures" instead of a single final architecture. Then during the inference phase, the controller selects an architecture from the distribution, tailored for each unseen image to achieve both high accuracy and short latency. The experimental results show that InstaNAS reduces the inference latency without compromising classification accuracy. On average, InstaNAS achieves 48.9% latency reduction on CIFAR-10 and 40.2% latency reduction on CIFAR-100 with respect to MobileNetV2 architecture.

Large mini-batch parallel SGD is commonly used for distributed training of deep networks. Approaches that use tightly-coupled exact distributed averaging based on AllReduce are sensitive to slow nodes and high-latency communication. In this work we show the applicability of Stochastic Gradient Push (SGP) for distributed training. SGP uses a gossip algorithm called PushSum for approximate distributed averaging, allowing for much more loosely coupled communications, which can be beneficial in high-latency or high-variability scenarios. The tradeoff is that approximate distributed averaging injects additional noise in the gradient which can affect the train and test accuracies. We prove that SGP converges to a stationary point of smooth, non-convex objective functions. Furthermore, we validate empirically the potential of SGP. For example, using 32 nodes with 8 GPUs per node to train ResNet-50 on ImageNet, where nodes communicate over 10Gbps Ethernet, SGP completes 90 epochs in around 1.6 hours while AllReduce SGD takes over 5 hours, and the top-1 validation accuracy of SGP remains within 1.2% of that obtained using AllReduce SGD.

DynamicGEM is an open-source Python library for learning node representations of dynamic graphs. It consists of state-of-the-art algorithms for defining embeddings of nodes whose connections evolve over time. The library also contains the evaluation framework for four downstream tasks on the network: graph reconstruction, static and temporal link prediction, node classification, and temporal visualization. We have implemented various metrics to evaluate the state-of-the-art methods, and examples of evolving networks from various domains. We have easy-to-use functions to call and evaluate the methods and have extensive usage documentation. Furthermore, DynamicGEM provides a template to add new algorithms with ease to facilitate further research on the topic.

Dept of Radiology, Geisinger, Danville, PA, United States Summary Predicting future clinical events, such as death, is an important task in medicine that helps physicians guide appropriate action. Neural networks have particular promise to assist with medical prediction tasks related to clinical imaging by learning patterns from large datasets. Moreover, neural networks have not yet been applied to medical videos on a large scale, such as ultrasound of the heart (echocardiography). Here we show that a large dataset of 723,754 clinically-acquired echocardiographic videos ( 45 million images) linked to longitudinal follow- up data in 27,028 patients can be used to train a deep neural network to predict 1-year survival with good accuracy. We also demonstrate that prediction accuracy can be further improved by adding highly predictive clinical variables from the electronic health record. Finally, in a blinded, independent test set, the trained neural network was more accurate in discriminating 1-year survival outcomes than two expert cardiologists. These results therefore highlight the potential of neural networks to add new predictive power to clinical image interpretations.

Diagnostic reasoning is a key component of many professions. To improve students' diagnostic reasoning skills, educational psychologists analyse and give feedback on epistemic activities used by these students while diagnosing, in particular, hypothesis generation, evidence generation, evidence evaluation, and drawing conclusions. However, this manual analysis is highly time-consuming. We aim to enable the large-scale adoption of diagnostic reasoning analysis and feedback by automating the epistemic activity identification. We create the first corpus for this task, comprising diagnostic reasoning self-explanations of students from two domains annotated with epistemic activities. Based on insights from the corpus creation and the task's characteristics, we discuss three challenges for the automatic identification of epistemic activities using AI methods: the correct identification of epistemic activity spans, the reliable distinction of similar epistemic activities, and the detection of overlapping epistemic activities. We propose a separate performance metric for each challenge and thus provide an evaluation framework for future research. Indeed, our evaluation of various state-of-the-art recurrent neural network architectures reveals that current techniques fail to address some of these challenges.

The meaning of a sentence is a function of the relations that hold between its words. We instantiate this relational view of semantics in a series of neural models based on variants of relation networks (RNs) which represent a set of objects (for us, words forming a sentence) in terms of representations of pairs of objects. We propose two extensions to the basic RN model for natural language. First, building on the intuition that not all word pairs are equally informative about the meaning of a sentence, we use constraints based on both supervised and unsupervised dependency syntax to control which relations influence the representation. Second, since higher-order relations are poorly captured by a sum of pairwise relations, we use a recurrent extension of RNs to propagate information so as to form representations of higher order relations. Experiments on sentence classification, sentence pair classification, and machine translation reveal that, while basic RNs are only modestly effective for sentence representation, recurrent RNs with latent syntax are a reliably powerful representational device.

Convolutional networks were initially developed for supervised learning. They are used in deep learning to classify two-dimensional spatial information such as hand writing samples and photographs [16]. In the one dimensional setting, they have been applied to temporal data such as audio recordings of speech and music, and writing encoded at either the character level or the word level. In the three dimensional setting, applications have included medical scans, object detection for self driving cars, and object recognition from RGB-D photos. Videos, with their two spatial dimensions and one time dimension can also be seen as 2 1 3 dimensional objects for purposes of applying convolutional networks [29]. The movement of 3D objects happens in 3 1 4 dimensional space-time, but 4D ConvNets are relatively unexplored.

Many dialogue management frameworks allow the system designer to directly define belief rules to implement an efficient dialog policy. Because these rules are directly defined, the components are said to be hand-crafted. As dialogues become more complex, the number of states, transitions, and policy decisions becomes very large. To facilitate the dialog policy design process, we propose an approach to automatically learn belief rules using a supervised machine learning approach. We validate our ideas in Student-Advisor conversation domain, where we extract latent beliefs like student is \textit{curious, confused and neutral}, etc. Further, we also perform epistemic reasoning that helps to tailor the dialog according to student's emotional state and hence improve the overall effectiveness of the dialog system. Our latent belief identification approach shows an accuracy of 87\% and this results in efficient and meaningful dialog management.

Deep learning based computer vision fails to work when labeled images are scarce. Recently, Meta learning algorithm has been confirmed as a promising way to improve the ability of learning from few images for computer vision. However, previous Meta learning approaches expose problems: 1) they ignored the importance of attention mechanism for the Meta learner; 2) they didn't give the Meta learner the ability of well using the past knowledge which can help to express images into high representations, resulting in that the Meta learner has to solve few shot learning task directly from the original high dimensional RGB images. In this paper, we argue that the attention mechanism and the past knowledge are crucial for the Meta learner, and the Meta learner should be trained on high representations of the RGB images instead of directly on the original ones. Based on these arguments, we propose two methods: Attention augmented Meta Learning (AML) and Representation based and Attention augmented Meta Learning(RAML). The method AML aims to improve the Meta learner's attention ability by explicitly embedding an attention model into its network. The method RAML aims to give the Meta learner the ability of leveraging the past learned knowledge to reduce the dimension of the original input data by expressing it into high representations, and help the Meta learner to perform well. Extensive experiments demonstrate the effectiveness of the proposed models, with state-of-the-art few shot learning performances on several few shot learning benchmarks. The source code of our proposed methods will be released soon to facilitate further studies on those aforementioned problem.

In this episode of our AI Platforms series, we're joined by Daniel Jeavons, General Manager of Data Science at Shell. In our conversation, Daniel and I explore the evolution of analytics and data science at Shell, and cover a ton of interesting machine learning use cases that the company is pursuing, such as well drilling and charging smart cars. A good bit of our conversation centers around IoT-related applications and issues, such as inference at the edge, federated machine learning, and digital twins, all key considerations for the way they apply ML. We also talk about the data science process at Shell and the importance of platform technologies to Daniel's organization and the company as a whole and we discuss some of the technologies he and his team are excited about introducing to the company. As many of you know, part of my work involves understanding the way large companies are adopting machine learning, deep learning and AI.