Generating an image from its description is a challenging task worth solving because of its numerous practical applications ranging from image editing to virtual reality. All existing methods use one single caption to generate a plausible image. A single caption by itself, can be limited, and may not be able to capture the variety of concepts and behavior that may be present in the image. We propose two deep generative models that generate an image by making use of multiple captions describing it. This is achieved by ensuring 'Cross-Caption Cycle Consistency' between the multiple captions and the generated image(s). We report quantitative and qualitative results on the standard Caltech-UCSD Birds (CUB) and Oxford-102 Flowers datasets to validate the efficacy of the proposed approach.

Traditional human activity recognition (HAR) based on time series adopts sliding window analysis method. This method faces the multi-class window problem which mistakenly labels different classes of sampling points within a window as a class. In this paper, a HAR algorithm based on U-Net is proposed to perform activity labeling and prediction at each sampling point. The activity data of the triaxial accelerometer is mapped into an image with the single pixel column and multi-channel which is input into the U-Net network for training and recognition. Our proposal can complete the pixel-level gesture recognition function. The method does not need manual feature extraction and can effectively identify short-term behaviors in long-term activity sequences. We collected the Sanitation dataset and tested the proposed scheme with four open data sets. The experimental results show that compared with Support Vector Machine (SVM), k-Nearest Neighbor (kNN), Decision Tree(DT), Quadratic Discriminant Analysis (QDA), Convolutional Neural Network (CNN) and Fully Convolutional Networks (FCN) methods, our proposal has the highest accuracy and F1-socre in each dataset, and has stable performance and high robustness. At the same time, after the U-Net has finished training, our proposal can achieve fast enough recognition speed.

The hypothesis that computational models can be reliable enough to be adopted in prognosis and patient care is revolutionizing healthcare. Deep learning, in particular, has been a game changer in building predictive models, thereby leading to community-wide data curation efforts. However, due to the inherent variabilities in population characteristics and biological systems, these models are often biased to the training datasets. This can be limiting when models are deployed in new environments, particularly when there are systematic domain shifts not known a priori. In this paper, we formalize these challenges by emulating a large class of domain shifts that can occur in clinical settings, and argue that evaluating the behavior of predictive models in light of those shifts is an effective way of quantifying the reliability of clinical models. More specifically, we develop an approach for building challenging scenarios, based on analysis of \textit{disease landscapes}, and utilize unsupervised domain adaptation to compensate for the domain shifts. Using the openly available MIMIC-III EHR dataset for phenotyping, we generate a large class of scenarios and evaluate the ability of deep clinical models in those cases. For the first time, our work sheds light into data regimes where deep clinical models can fail to generalize, due to significant changes in the disease landscapes between the source and target landscapes. This study emphasizes the need for sophisticated evaluation mechanisms driven by real-world domain shifts to build effective AI solutions for healthcare.

In many classification problems it is desirable to output well-calibrated probabilities on the different classes. We propose a robust, non-parametric method of calibrating probabilities called SplineCalib that utilizes smoothing splines to determine a calibration function. We demonstrate how applying certain transformations as part of the calibration process can improve performance on problems in deep learning and other domains where the scores tend to be "overconfident". We adapt the approach to multi-class problems and find that better calibration can improve accuracy as well as log-loss by better resolving uncertain cases. Finally, we present a cross-validated approach to calibration which conserves data. Significant improvements to log-loss and accuracy are shown on several different problems. We also introduce the ml-insights python package which contains an implementation of the SplineCalib algorithm.

Through many recent successes in simulation, model-free reinforcement learning has emerged as a promising approach to solving continuous control robotic tasks. The research community is now able to reproduce, analyze and build quickly on these results due to open source implementations of learning algorithms and simulated benchmark tasks. To carry forward these successes to real-world applications, it is crucial to withhold utilizing the unique advantages of simulations that do not transfer to the real world and experiment directly with physical robots. However, reinforcement learning research with physical robots faces substantial resistance due to the lack of benchmark tasks and supporting source code. In this work, we introduce several reinforcement learning tasks with multiple commercially available robots that present varying levels of learning difficulty, setup, and repeatability. On these tasks, we test the learning performance of off-the-shelf implementations of four reinforcement learning algorithms and analyze sensitivity to their hyper-parameters to determine their readiness for applications in various real-world tasks. Our results show that with a careful setup of the task interface and computations, some of these implementations can be readily applicable to physical robots. We find that state-of-the-art learning algorithms are highly sensitive to their hyper-parameters and their relative ordering does not transfer across tasks, indicating the necessity of re-tuning them for each task for best performance. On the other hand, the best hyper-parameter configuration from one task may often result in effective learning on held-out tasks even with different robots, providing a reasonable default. We make the benchmark tasks publicly available to enhance reproducibility in real-world reinforcement learning.

This paper proposes a simple, yet very effective method to localize dominant foreground objects in an image, to pixel-level precision. The proposed method 'MASON' (Model-AgnoStic ObjectNess) uses a deep convolutional network to generate category-independent and model-agnostic heat maps for any image. The network is not explicitly trained for the task, and hence, can be used off-the-shelf in tandem with any other network or task. We show that this framework scales to a wide variety of images, and illustrate the effectiveness of MASON in three varied application contexts.

You visit your doctor for a routine checkup. She decides to do a fine needle aspiration for a seemingly insignificant lesion ("It's probably nothing…"). But a few days later you get a call back. They've found some abnormal cells. Your doctor recommends a specialist who performs a resection in his office, sending a tissue sample to a histology lab.

From smartphone assistants to image recognition and translation, machine learning already helps us in our everyday lives. But it can also help us to tackle some of the world's most challenging physical problems -- such as energy consumption. Large-scale commercial and industrial systems like data centres consume a lot of energy, and while much has been done to stem the growth of energy use, there remains a lot more to do given the world's increasing need for computing power. Reducing energy usage has been a major focus for us over the past 10 years: we have built our own super-efficient servers at Google, invented more efficient ways to cool our data centres and invested heavily in green energy sources, with the goal of being powered 100 percent by renewable energy. Compared to five years ago, we now get around 3.5 times the computing power out of the same amount of energy, and we continue to make many improvements each year.

" I'm intrigued by the general trend towards empowering individuals to share their data in a secure and controlled environment. Democratisation of data in this way has to be the future. Imagine what we will be able to do in decades to come, when individuals have access to their complete healthcare records in electronic form, paired with high quality data from genomics, epigenetics, microbiome, imaging, activity and lifestyle profiles, etc., supported by a platform that enables individuals to share all or parts of their data with partners of their choice, for purposes they care about, in return for services they value – very exciting! I have interviewed Bryn Roberts, Global Head of Operations for Roche Pharmaceutical Research & Early Development, and Site Head in Basel. We talked about using AI and Data Analytics in Pharmaceutical Research.

In 1976, Maxmen1 predicted that artificial intelligence (AI) in the 21st century would usher in "the post-physician era," with health care provided by paramedics and computers. Today, the mass extinction of physicians remains unlikely. However, as outlined by Hinton2 in a related Viewpoint, the emergence of a radically different approach to AI, called deep learning, has the potential to effect major changes in clinical medicine and health care delivery.