Putting the user first is the approach that Fast.ai took when building their Deep Learning library. I think of Jeremy Howard as the Don Norman of Machine Learning. Instead of just focusing on the model building part, Fast.ai builds tools around all of the below. By tools I don't mean a black box service, I mean software design patterns specific to Machine Learning.

Fabrizio Augusto Poltronieri is an award-winning computer artist working with Creative AI, exploring the relationship between technology and deep-rooted philosophical concepts, such as chance. His current artwork involves Artificial Intelligence, applying machine and deep learning techniques to create and design narratives, moving images and objects. He started his artistic career as the founder of the art group [ zero], in 2006, which for over 13 years has participated in international exhibitions and festivals such as the Ars Electronica, in Linz, Austria, presenting artworks and live performances that plays with the relationship man-machine. With his solo artistic practice, Poltronieri became an award-winning artist, with a long list of exhibitions. His artworks are in prominent collections such as that of the Victoria & Albert Museum in London.

A deep learning model successfully predicted the lung cancer survival period with an accuracy of 71.18%, outperforming previous machine learning models, according to the results of a study published in the International Journal of Medical Informatics. "Early detection and prediction of depth of survivability from cancer can help both patients and healthcare professionals better manage costs, treatment intensity and time spent around medical care," the authors wrote. The aim of this study was to characterize a deep learning approach to predict the survival period of patients with lung cancer. The study used data from the Surveillance, Epidemiology, and End Results (SEER) program. The deep learning models included data preprocessing using categorical and quantitative variables.

Reading text in natural images has attracted increasing attention in the computer vision community due to its numerous practical applications in document analysis, scene understanding, robot navigation, and image retrieval. Although previous works have made significant progress in both text detection and text recognition, it is still challenging due to the large variance of text patterns and highly complicated background. The most common way in scene text reading is to divide it into text detection and text recognition, which are handled as two separate tasks. Deep learning based approaches become dominate in both parts. Text Detection: Text Detection is a technique where image will be given to the model and the textual region is detected by plotting a bounding box around it.

Artificial intelligence (AI) is the ultimate manifestation of the promise of technology to improve our lives. Within our lifetimes AI will become omnipresent across every segment of economic activity and within every field of human endeavour. But AI is not a single genus but a constellation of technologies that must work together in harmony. By 2030 over 50 billion devices will be connected to the Internet of Things and will include new sensor technologies which will capture and create new modalities of data. Mastering this constellation of artificial intelligence technologies is a rainbow of minds from data scientists to robot builders and AI investors.

In this paper, we focus on the problem of unsupervised image-sentence matching. Existing research explores to utilize document-level structural information to sample positive and negative instances for model training. Although the approach achieves positive results, it introduces a sampling bias and fails to distinguish instances with high semantic similarity. To alleviate the bias, we propose a new sampling strategy to select additional intra-document image-sentence pairs as positive or negative samples. Furthermore, to recognize the complex pattern in intra-document samples, we propose a Transformer based model to capture fine-grained features and implicitly construct a graph for each document, where concepts in a document are introduced to bridge the representation learning of images and sentences in the context of a document. Experimental results show the effectiveness of our approach to alleviate the bias and learn well-aligned multimodal representations.

In this paper, we propose a novel lightweight relation extraction approach of structural block driven - convolutional neural learning. Specifically, we detect the essential sequential tokens associated with entities through dependency analysis, named as a structural block, and only encode the block on a block-wise and an inter-block-wise representation, utilizing multi-scale CNNs. This is to 1) eliminate the noisy from irrelevant part of a sentence; meanwhile 2) enhance the relevant block representation with both block-wise and inter-block-wise semantically enriched representation. Our method has the advantage of being independent of long sentence context since we only encode the sequential tokens within a block boundary. Experiments on two datasets i.e., SemEval2010 and KBP37, demonstrate the significant advantages of our method. In particular, we achieve the new state-of-the-art performance on the KBP37 dataset; and comparable performance with the state-of-the-art on the SemEval2010 dataset.

We focus on the robustness of neural networks for classification. To permit a fair comparison between methods to achieve robustness, we first introduce a standard based on the mensuration of a classifier's degradation. Then, we propose natural perturbed training to robustify the network. Natural perturbations will be encountered in practice: the difference of two images of the same object may be approximated by an elastic deformation (when they have slightly different viewing angles), by occlusions (when they hide differently behind objects), or by saturation, Gaussian noise etc. Training some fraction of the epochs on random versions of such variations will help the classifier to learn better. We conduct extensive experiments on six datasets of varying sizes and granularity. Natural perturbed learning show better and much faster performance than adversarial training on clean, adversarial as well as natural perturbed images. It even improves general robustness on perturbations not seen during the training. For Cifar-10 and STL-10 natural perturbed training even improves the accuracy for clean data and reaches the state of the art performance. Ablation studies verify the effectiveness of natural perturbed training.

We introduce a novel and highly tractable supervised learning approach based on neural networks that can be applied for the computation of model-free price bounds of, potentially high-dimensional, financial derivatives and for the determination of optimal hedging strategies attaining these bounds. In particular, our methodology allows to train a single neural network offline and then to use it online for the fast determination of model-free price bounds of a whole class of financial derivatives with current market data. We show the applicability of this approach and highlight its accuracy in several examples involving real market data. Further, we show how a neural network can be trained to solve martingale optimal transport problems involving fixed marginal distributions instead of financial market data.

Relocalization is a fundamental task in the field of robotics and computer vision. There is considerable work in the field of deep camera relocalization, which directly estimates poses from raw images. However, learning-based methods have not yet been applied to the radar sensory data. In this work, we investigate how to exploit deep learning to predict global poses from Emerging Frequency-Modulated Continuous Wave (FMCW) radar scans. Specifically, we propose a novel end-to-end neural network with self-attention, termed RadarLoc, which is able to estimate 6-DoF global poses directly. We also propose to improve the localization performance by utilizing geometric constraints between radar scans. We validate our approach on the recently released challenging outdoor dataset Oxford Radar RobotCar. Comprehensive experiments demonstrate that the proposed method outperforms radar-based localization and deep camera relocalization methods by a significant margin.