Catchy Christmas songs can now be created by a special songwriting AI, taught by studying existing festive tunes. The system came up with catchy jingles with names like'Syllabub Chocolatebell', 'Peaches Twinkleleaves' and'Cocoa Jollyfluff'. Researchers from Made by AI trained a neural network by inputting one hundred Christmas tunes in the form of Musical Instrument Digital Interface (MIDI) files. It then picked out recurring themes, motifs, instruments and rhythms to generate its own hits. Scientists have trained an AI system to write its own catchy Christmas songs by teaching it existing festive tunes.

We've heard from customers that scaling TensorFlow training jobs to multiple nodes and GPUs successfully is hard. TensorFlow has distributed training built-in, but it can be difficult to use. Recently, we made optimizations to TensorFlow and Horovod to help AWS customers scale TensorFlow training jobs to multiple nodes and GPUs. With these improvements, any AWS customer can use an AWS Deep Learning AMI to train ResNet-50 on ImageNet in just under 15 minutes. To achieve this, 32 Amazon EC2 instances, each with 8 GPUs, a total 256 GPUs, were harnessed with TensorFlow. All of the required software and tools for this solution ship with the latest Deep Learning AMIs (DLAMIs), so you can try it out yourself. You can train faster, implement your models faster, and get results faster than ever before. This blog post describes our results and shows you how to try out this easier and faster way to run distributed training with TensorFlow. Figure A. ResNet-50 ImageNet model training with the latest optimized TensorFlow with Horovod on a Deep Learning AMI takes 15 minutes on 256 GPUs.

The objective of a neural network is to have a final model that performs well both on the data that we used to train it (e.g. the training dataset) and the new data on which the model will be used to make predictions. The central challenge in machine learning is that we must perform well on new, previously unseen inputs -- not just those on which our model was trained. The ability to perform well on previously unobserved inputs is called generalization.

The literature about history matching is vast and despite the impressive number of methods proposed and the significant progresses reported in the last decade, conditioning reservoir models to dynamic data is still a challenging task. Ensemble-based methods are among the most successful and efficient techniques currently available for history matching. These methods are usually able to achieve reasonable data matches, especially if an iterative formulation is employed. However, they sometimes fail to preserve the geological realism of the model, which is particularly evident in reservoir with complex facies distributions. This occurs mainly because of the Gaussian assumptions inherent in these methods. This fact has encouraged an intense research activity to develop parameterizations for facies history matching. Despite the large number of publications, the development of robust parameterizations for facies remains an open problem. Deep learning techniques have been delivering impressive results in a number of different areas and the first applications in data assimilation in geoscience have started to appear in literature. The present paper reports the current results of our investigations on the use of deep neural networks towards the construction of a continuous parameterization of facies which can be used for data assimilation with ensemble methods. Specifically, we use a convolutional variational autoencoder and the ensemble smoother with multiple data assimilation. We tested the parameterization in three synthetic history-matching problems with channelized facies. We focus on this type of facies because they are among the most challenging to preserve after the assimilation of data. The parameterization showed promising results outperforming previous methods and generating well-defined channelized facies.

Crowd counting problem aims to count the number of objects within an image or a frame in the videos and is usually solved by estimating the density map generated from the object location annotations. The values in the density map, by nature, take two possible states: zero indicating no object around, a non-zero value indicating the existence of objects and the value denoting the local object density. In contrast to traditional methods which do not differentiate the density prediction of these two states, we propose to use a dedicated network branch to predict the object/non-object mask and then combine its prediction with the input image to produce the density map. Our rationale is that the mask prediction could be better modeled as a binary segmentation problem and the difficulty of estimating the density could be reduced if the mask is known. A key to the proposed scheme is the strategy of incorporating the mask prediction into the density map estimator. To this end, we study five possible solutions, and via analysis and experimental validation we identify the most effective one. Through extensive experiments on five public datasets, we demonstrate the superior performance of the proposed approach over the baselines and show that our network could achieve the state-of-the-art performance.

We propose a new deep learning approach for medical imaging that copes with the problem of a small training set, the main bottleneck of deep learning, and apply it for classification of healthy and cancer cells acquired by quantitative phase imaging. The proposed method, called transferring of pre-trained generative adversarial network (TOP-GAN), is a hybridization between transfer learning and generative adversarial networks (GANs). Healthy cells and cancer cells of different metastatic potential have been imaged by low-coherence off-axis holography. After the acquisition, the optical path delay maps of the cells have been extracted and directly used as an input to the deep networks. In order to cope with the small number of classified images, we have used GANs to train a large number of unclassified images from another cell type (sperm cells). After this preliminary training, and after transforming the last layer of the network with new ones, we have designed an automatic classifier for the correct cell type (healthy/primary cancer/metastatic cancer) with 90-99% accuracy, although small training sets of down to several images have been used. These results are better in comparison to other classic methods that aim at coping with the same problem of a small training set. We believe that our approach makes the combination of holographic microscopy and deep learning networks more accessible to the medical field by enabling a rapid, automatic and accurate classification in stain-free imaging flow cytometry. Furthermore, our approach is expected to be applicable to many other medical image classification tasks, suffering from a small training set.

Learning from data in the quaternion domain enables us to exploit internal dependencies of 4D signals and treating them as a single entity. One of the models that perfectly suits with quaternion-valued data processing is represented by 3D acoustic signals in their spherical harmonics decomposition. In this paper, we address the problem of localizing and detecting sound events in the spatial sound field by using quaternion-valued data processing. In particular, we consider the spherical harmonic components of the signals captured by a first-order ambisonic microphone and process them by using a quaternion convolutional neural network. Experimental results show that the proposed approach exploits the correlated nature of the ambisonic signals, thus improving accuracy results in 3D sound event detection and localization.

In this paper, we provide a modern synthesis of the classic inverse compositional algorithm for dense image alignment. We first discuss the assumptions made by this well-established technique, and subsequently propose to relax these assumptions by incorporating data-driven priors into this model. More specifically, we unroll a robust version of the inverse compositional algorithm and replace multiple components of this algorithm using more expressive models whose parameters we train in an end-to-end fashion from data. Our experiments on several challenging 3D rigid motion estimation tasks demonstrate the advantages of combining optimization with learning-based techniques, outperforming the classic inverse compositional algorithm as well as data-driven image-to-pose regression approaches.

In this paper, we address the problem of synthesizing multi-parameter magnetic resonance imaging (mp-MRI) data, i.e. Apparent Diffusion Coefficients (ADC) and T2-weighted (T2w), containing clinically significant (CS) prostate cancer (PCa) via semi-supervised adversarial learning. Specifically, our synthesizer generates mp-MRI data in a sequential manner: first generating ADC maps from 128-d latent vectors, followed by translating them to the T2w images. The synthesizer is trained in a semisupervised manner. In the supervised training process, a limited amount of paired ADC-T2w images and the corresponding ADC encodings are provided and the synthesizer learns the paired relationship by explicitly minimizing the reconstruction losses between synthetic and real images. To avoid overfitting limited ADC encodings, an unlimited amount of random latent vectors and unpaired ADC-T2w Images are utilized in the unsupervised training process for learning the marginal image distributions of real images. To improve the robustness of synthesizing, we decompose the difficult task of generating full-size images into several simpler tasks which generate sub-images only. A StitchLayer is then employed to fuse sub-images together in an interlaced manner into a full-size image. To enforce the synthetic images to indeed contain distinguishable CS PCa lesions, we propose to also maximize an auxiliary distance of Jensen-Shannon divergence (JSD) between CS and nonCS images. Experimental results show that our method can effectively synthesize a large variety of mpMRI images which contain meaningful CS PCa lesions, display a good visual quality and have the correct paired relationship. Compared to the state-of-the-art synthesis methods, our method achieves a significant improvement in terms of both visual and quantitative evaluation metrics.

In neural networks, it is often desirable to work with various representations of the same space. For example, 3D rotations can be represented with quaternions or Euler angles. In this paper, we advance a definition of a continuous representation, which can be helpful for training deep neural network. We relate this to the definition of topological equivalence. We then investigate what are continuous and discontinuous representations for 2D, 3D, and n-dimensional rotations. We demonstrate that for 3D rotations, all representations are discontinuous in four or fewer dimensions in real Euclidean space. Thus, widely used representations such as quaternions and Euler angles are discontinuous and difficult for neural networks to learn. We show that the 3D rotations have continuous representations in 5D and 6D which are more suitable for learning. We also present continuous representations for the general case of the n dimensional rotation group SO(n). While our main focus is on rotations, we also show that our constructions apply to other groups such as the orthogonal group and similarity transforms. We finally present empirical results, which show that our continuous rotation representations outperform discontinuous ones for several practical problems in graphics and vision, including a simple autoencoder sanity test, a rotation estimator for 3D point clouds, and an inverse kinematics solver for 3D human poses.