The CP decomposition for high dimensional non-orthogonal spike tensors is an important problem with broad applications across many disciplines. However, previous works with theoretical guarantee typically assume restrictive incoherence conditions on the basis vectors for the CP components. In this paper, we propose new computationally efficient composite PCA and concurrent orthogonalization algorithms for tensor CP decomposition with theoretical guarantees under mild incoherence conditions. The composite PCA applies the principal component or singular value decompositions twice, first to a matrix unfolding of the tensor data to obtain singular vectors and then to the matrix folding of the singular vectors obtained in the first step. It can be used as an initialization for any iterative optimization schemes for the tensor CP decomposition. The concurrent orthogonalization algorithm iteratively estimates the basis vector in each mode of the tensor by simultaneously applying projections to the orthogonal complements of the spaces generated by others CP components in other modes. It is designed to improve the alternating least squares estimator and other forms of the high order orthogonal iteration for tensors with low or moderately high CP ranks. Our theoretical investigation provides estimation accuracy and statistical convergence rates for the two proposed algorithms. Our implementations on synthetic data demonstrate significant practical superiority of our approach over existing methods.

Big Data are rapidly produced from various heterogeneous data sources. They are of different types (text, image, video or audio) and have different levels of reliability and completeness. One of the most interesting architectures that deal with the large amount of emerging data at high velocity is called the lambda architecture. In fact, it combines two different processing layers namely batch and speed layers, each providing specific views of data while ensuring robustness, fast and scalable data processing. However, most papers dealing with the lambda architecture are focusing one single type of data generally produced by a single data source. Besides, the layers of the architecture are implemented independently, or, at best, are combined to perform basic processing without assessing either the data reliability or completeness. Therefore, inspired by the lambda architecture, we propose in this paper a generic multimodal architecture that combines both batch and streaming processing in order to build a complete, global and accurate insight in near-real-time based on the knowledge extracted from multiple heterogeneous Big Data sources. Our architecture uses batch processing to analyze the data structures and contents, build the learning models and calculate the reliability index of the involved sources, while the streaming processing uses the built-in models of the batch layer to immediately process incoming data and rapidly provide results. We validate our architecture in the context of urban traffic management systems in order to detect congestions.

Natural numerical networks are introduced as a new classification algorithm based on the numerical solution of nonlinear partial differential equations of forward-backward diffusion type on complete graphs. The proposed natural numerical network is applied to open important environmental and nature conservation task, the automated identification of protected habitats by using satellite images. In the natural numerical network, the forward diffusion causes the movement of points in a feature space toward each other. The opposite effect, keeping the points away from each other, is caused by backward diffusion. This yields the desired classification. The natural numerical network contains a few parameters that are optimized in the learning phase of the method. After learning parameters and optimizing the topology of the network graph, classification necessary for habitat identification is performed. A relevancy map for each habitat is introduced as a tool for validating the classification and finding new Natura 2000 habitat appearances.

The performance of deep neural networks typically increases with the number of training images. However, not all images have the same importance towards improved performance and robustness. In fetal brain MRI, abnormalities exacerbate the variability of the developing brain anatomy compared to non-pathological cases. A small number of abnormal cases, as is typically available in clinical datasets used for training, are unlikely to fairly represent the rich variability of abnormal developing brains. This leads machine learning systems trained by maximizing the average performance to be biased toward non-pathological cases. This problem was recently referred to as hidden stratification. To be suited for clinical use, automatic segmentation methods need to reliably achieve high-quality segmentation outcomes also for pathological cases. In this paper, we show that the state-of-the-art deep learning pipeline nnU-Net has difficulties to generalize to unseen abnormal cases. To mitigate this problem, we propose to train a deep neural network to minimize a percentile of the distribution of per-volume loss over the dataset. We show that this can be achieved by using Distributionally Robust Optimization (DRO). DRO automatically reweights the training samples with lower performance, encouraging nnU-Net to perform more consistently on all cases. We validated our approach using a dataset of 368 fetal brain T2w MRIs, including 124 MRIs of open spina bifida cases and 51 MRIs of cases with other severe abnormalities of brain development.

The goals of this paper are two-fold. The first goal is to serve as an expository tutorial on the working of deep learning models which emphasizes geometrical intuition about the reasons for success of deep learning. The second goal is to complement the current results on the expressive power of deep learning models and their loss surfaces with novel insights and results. In particular, we describe how deep neural networks carve out manifolds especially when the multiplication neurons are introduced. Multiplication is used in dot products and the attention mechanism and it is employed in capsule networks and self-attention based transformers. We also describe how random polynomial, random matrix, spin glass and computational complexity perspectives on the loss surfaces are interconnected.

Along with climate change, more frequent extreme events, such as flooding and tropical cyclones, threaten the livelihoods and wellbeing of poor and vulnerable populations. One of the most immediate needs of people affected by a disaster is finding shelter. While the proliferation of data on disasters is already helping to save lives, identifying damages in buildings, assessing shelter needs, and finding appropriate places to establish emergency shelters or settlements require a wide range of data to be combined rapidly. To address this gap and make a headway in comprehensive assessments, this paper proposes a machine learning workflow that aims to fuse and rapidly analyse multimodal data. This workflow is built around open and online data to ensure scalability and broad accessibility. Based on a database of 19 characteristics for more than 200 disasters worldwide, a fusion approach at the decision level was used. This technique allows the collected multimodal data to share a common semantic space that facilitates the prediction of individual variables. Each fused numerical vector was fed into an unsupervised clustering algorithm called Self-Organizing-Maps (SOM). The trained SOM serves as a predictor for future cases, allowing predicting consequences such as total deaths, total people affected, and total damage, and provides specific recommendations for assessments in the shelter and housing sector. To achieve such prediction, a satellite image from before the disaster and the geographic and demographic conditions are shown to the trained model, which achieved a prediction accuracy of 62 %

Neural text generation models are typically trained by maximizing log-likelihood with the sequence cross entropy loss, which encourages an exact token-by-token match between a target sequence with a generated sequence. Such training objective is sub-optimal when the target sequence not perfect, e.g., when the target sequence is corrupted with noises, or when only weak sequence supervision is available. To address this challenge, we propose a novel Edit-Invariant Sequence Loss (EISL), which computes the matching loss of a target n-gram with all n-grams in the generated sequence. EISL draws inspirations from convolutional networks (ConvNets) which are shift-invariant to images, hence is robust to the shift of n-grams to tolerate edits in the target sequences. Moreover, the computation of EISL is essentially a convolution operation with target n-grams as kernels, which is easy to implement with existing libraries. To demonstrate the effectiveness of EISL, we conduct experiments on three tasks: machine translation with noisy target sequences, unsupervised text style transfer, and non-autoregressive machine translation. Experimental results show our method significantly outperforms cross entropy loss on these three tasks.

Artificial intelligence (AI) is already embedded in a range of digital services. Voice assistants such as Alexa, car routing or content translation all involve machine learning - the most popular form of artificial intelligence technology. There are many warnings these days about AI, such as the ethics behind these machine driven decision systems or threats of automation and the loss of many jobs. Very little is reported about how artificial intelligence can improve public services and can have positive social impact. Smart algorithms combined with cloud computing power allow unprecedented forms of data analysis that would take much longer if humans were doing it.

At first blush, the idea of scanning images synced up to iCloud for child sexual abuse materials against the hash list of known CSAM images seems like a good idea. As a survivor of childhood sexual abuse myself, I want tech companies to takes some initiative to deal with this issue. They also want to scan images on kids' phones using AI to see if kids are getting into any trouble with sending or receiving sexual material. Again, that sounds like a good thing. But, as the EFF points out, this all requires a backdoor, and backdoors, once created, almost never remain used for just one purpose.

A new report from FICO and Corinium has found that many companies are deploying various forms of AI throughout their businesses with little consideration for the ethical implications of potential problems. The increasing scale of AI is raising the stakes for major ethical questions. There have been hundreds of examples over the last decade of the many disastrous ways AI has been used by companies, from facial recognition systems unable to discern darker skinned faces to healthcare apps that discriminate against African American patients to recidivism calculators used by courts that skew against certain races. Despite these examples, FICO's State of Responsible AI report shows business leaders are putting little effort into ensuring that the AI systems they use are both fair and safe for widespread use. The survey, conducted in February and March, features the insights of 100 AI-focused leaders from the financial services sector, with 20 executives hailing from the US, Latin America, Europe, the Middle East, Africa, and the Asia Pacific regions.