This is the first chapter in Element AI's Foundations Series on AI-First Design (AI1D). Each chapter aims to define the component parts of AI1D in order to create a common language with which to explore this new era of design. You can read the intro to the series here, and sign up to stay tuned for the next chapter here. As a designer, why should you need to be able to understand artificial intelligence? It's a term being bandied about so much in media and tech circles lately, a kind of catchall that could be describing anything from virtual personal assistants, robots, sci-fi characters, or the latest deep learning algorithm. Perhaps you work in AI and you have a more nuanced understanding of these distinct fields, or maybe you just sense that your work will be affected in some way by AI in the coming years, but you're not quite sure how.

Booz Allen and NVIDIA are offering deep learning training. NVIDIA is working with Booz Allen Hamilton to rapidly build solutions that are needed in cyberdefense for both government and commercial customers. Now, certified Deep Learning Institute instructors from NVIDIA and Booz Allen are offering training to a variety of customers on how to build your own effective deep learning and data-driven solutions. 'Deep Learning Demystified,' hosted by Booz Allen and NVIDIA, will provide an introduction to deep learning, explore key fundamentals and opportunities, and how to best address current challenges. If you can't make our June 7th, 7:30AM - 11:00AM course, this course will also be offered over another two dates: "Together with NVIDIA, we've already seen through the Data Science Bowl how deep learning can speed cancer and heart disease diagnoses. Machine intelligence, powered by deep learning and other techniques, will help organizations -- public and private sector alike -- supercharge human ingenuity to uncover new revelations about the complex systems in which we live and work," said Dr. Josh Sullivan, senior vice president of data science at Booz Allen.

Inspired by the tremendous success of deep generative models on generating continuous data like image and audio, in the most recent year, few deep graph generative models have been proposed to generate discrete data such as graphs. They are typically unconditioned generative models which has no control on modes of the graphs being generated. Differently, in this paper, we are interested in a new problem named \emph{Deep Graph Translation}: given an input graph, we want to infer a target graph based on their underlying (both global and local) translation mapping. Graph translation could be highly desirable in many applications such as disaster management and rare event forecasting, where the rare and abnormal graph patterns (e.g., traffic congestions and terrorism events) will be inferred prior to their occurrence even without historical data on the abnormal patterns for this graph (e.g., a road network or human contact network). To achieve this, we propose a novel Graph-Translation-Generative Adversarial Networks (GT-GAN) which will generate a graph translator from input to target graphs. GT-GAN consists of a graph translator where we propose new graph convolution and deconvolution layers to learn the global and local translation mapping. A new conditional graph discriminator has also been proposed to classify target graphs by conditioning on input graphs. Extensive experiments on multiple synthetic and real-world datasets demonstrate the effectiveness and scalability of the proposed GT-GAN.

Few-shot learning aims to build a learner that quickly generalizes to novel classes even when a limited number of labeled examples (so-called low-data problem) are available. Meta-learning is commonly deployed to mimic the test environment in a training phase for good generalization, where episodes (i.e., learning problems) are manually constructed from the training set. This framework gains a lot of attention to few-shot learning with impressive performance, though the low-data problem is not fully addressed. In this paper, we propose Transductive Propagation Network (TPN), a transductive method that classifies the entire test set at once to alleviate the low-data problem. Specifically, our proposed network explicitly learns an underlying manifold space that is appropriate to propagate labels from few-shot examples, where all parameters of feature embedding, manifold structure, and label propagation are estimated in an end-to-end way on episodes. We evaluate the proposed method on the commonly used miniImageNet and tieredImageNet benchmarks and achieve the state-of-the-art or promising results on these datasets.

The robust and efficient recognition of visual relations in images is a hallmark of biological vision. We argue that, despite recent progress in visual recognition, modern machine vision algorithms are severely limited in their ability to learn visual relations. Through controlled experiments, we demonstrate that visual-relation problems strain convolutional neural networks (CNNs). The networks eventually break altogether when rote memorization becomes impossible, as when intra-class variability exceeds network capacity. Motivated by the comparable success of biological vision, we argue that feedback mechanisms including attention and perceptual grouping may be the key computational components underlying abstract visual reasoning.\

Although neural machine translation (NMT) with the encoder-decoder framework has achieved great success in recent times, it still suffers from some drawbacks: RNNs tend to forget old information which is often useful in the current step and the encoder only operates over words without considering word relationship. To solve these problems, we introduce relation networks (RNs) to learn better representations of the source. In our method RNs are used to associate source words with each other so that the source representation can memorize all the source words and also contain the relationship between them. Then the source representations and all the relations are fed into the attention component together while decoding, with the main encoder-decoder architecture unchanged. Experiments on several data sets show that our method can improve the translation performance significantly over the conventional encoder-decoder model, and can even outperform the approach involving supervised syntactic knowledge.

Large amounts of labeled data are typically required to train deep learning models. For many real-world problems, however, acquiring additional data can be expensive or even impossible. We present semi-supervised deep kernel learning (SSDKL), a semi-supervised regression model based on minimizing predictive variance in the posterior regularization framework. SSDKL combines the hierarchical representation learning of neural networks with the probabilistic modeling capabilities of Gaussian processes. By leveraging unlabeled data, we show improvements on a diverse set of real-world regression tasks over supervised deep kernel learning and semi-supervised methods such as VAT and mean teacher adapted for regression.

The predictive power of Convolutional Neural Networks (CNNs) has been an integral factor for emerging latency-sensitive applications, such as autonomous drones and vehicles. Such systems employ multiple CNNs, each one trained for a particular task. The efficient mapping of multiple CNNs on a single FPGA device is a challenging task as the allocation of compute resources and external memory bandwidth needs to be optimised at design time. This paper proposes f-CNN$^{\text{x}}$, an automated toolflow for the optimised mapping of multiple CNNs on FPGAs, comprising a novel multi-CNN hardware architecture together with an automated design space exploration method that considers the user-specified performance requirements for each model to allocate compute resources and generate a synthesisable accelerator. Moreover, f-CNN$^{\text{x}}$ employs a novel scheduling algorithm that alleviates the limitations of the memory bandwidth contention between CNNs and sustains the high utilisation of the architecture. Experimental evaluation shows that f-CNN$^{\text{x}}$'s designs outperform contention-unaware FPGA mappings by up to 50% and deliver up to 6.8x higher performance-per-Watt over highly optimised GPU designs for multi-CNN systems.

Generalization has been one of the major challenges for learning dynamics models in model-based reinforcement learning. However, previous work on action-conditioned dynamics prediction focuses on learning the pixel-level motion and thus does not generalize well to novel environments with different object layouts. In this paper, we present a novel object-oriented framework, called object-oriented dynamics predictor (OODP), which decomposes the environment into objects and predicts the dynamics of objects conditioned on both actions and object-to-object relations. It is an end-to-end neural network and can be trained in an unsupervised manner. To enable the generalization ability of dynamics learning, we design a novel CNN-based relation mechanism that is class-specific (rather than object-specific) and exploits the locality principle. Empirical results show that OODP significantly outperforms previous methods in terms of generalization over novel environments with various object layouts. OODP is able to learn from very few environments and accurately predict dynamics in a large number of unseen environments. In addition, OODP learns semantically and visually interpretable dynamics models.

Maps are an important medium that enable people to comprehensively understand the configuration of cultural activities and natural elements over different times and places. Although massive maps are available in the digital era, how to effectively and accurately access the required map remains a challenge today. Previous works partially related to map-type classification mainly focused on map comparison and map matching at the local scale. The features derived from local map areas might be insufficient to characterize map content. To facilitate establishing an automatic approach for accessing the needed map, this paper reports our investigation into using deep learning techniques to recognize seven types of map, including topographic map, terrain map, physical map, urban scene map, the National Map, 3D map, nighttime map, orthophoto map, and land cover classification map. Experimental results show that the state-of-the-art deep convolutional neural networks can support automatic map-type classification. Additionally, the classification accuracy varies according to different map-types. We hope our work can contribute to the implementation of deep learning techniques in cartographical community and advance the progress of Geographical Artificial Intelligence (GeoAI).