Using Artificial Intelligence to Augment Human Intelligence

Carter, Shan; Nielsen, Michael

doi:10.23915/distill.00009

Cited by 108 publications

(67 citation statements)

References 10 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…We demonstrate these results on a large dataset of fonts and highlight the limitations of a sequential, stochastic model for capturing the statistical dependencies and richness of this dataset. Even in its present form, the current model may be employed as an assistive agent for helping humans design fonts in a more time-efficient manner [6,46]. For example, a human may design a small set of characters and employ style propagation to synthesize the remaining set of characters ( Figure 5, 6).…”

Section: Discussionmentioning

confidence: 99%

A Learned Representation for Scalable Vector Graphics

Lopes

Eck

et al. 2019

2019 IEEE/CVF International Conference on Computer Vision (ICCV)

View full text Add to dashboard Cite

Dramatic advances in generative models have resulted in near photographic quality for artificially rendered faces, animals and other objects in the natural world. In spite of such advances, a higher level understanding of vision and imagery does not arise from exhaustively modeling an object, but instead identifying higher-level attributes that best summarize the aspects of an object. In this work we attempt to model the drawing process of fonts by building sequential generative models of vector graphics. This model has the benefit of providing a scale-invariant representation for imagery whose latent representation may be systematically manipulated and exploited to perform style propagation. We demonstrate these results on a large dataset of fonts and highlight how such a model captures the statistical dependencies and richness of this dataset. We envision that our model can find use as a tool for graphic designers to facilitate font design. * Work done as a member of the Google AI Residency Program (g. co/airesidency) Learned Vector Graphics RepresentationPixel Counterpart moveTo (15, 25) lineTo (-2, 0.3) cubicBezier

show abstract

Section: Discussionmentioning

confidence: 99%

A Learned Representation for Scalable Vector Graphics

Lopes

Eck

et al. 2019

2019 IEEE/CVF International Conference on Computer Vision (ICCV)

View full text Add to dashboard Cite

show abstract

“…Our approach is an example of artificial intelligence augmentation [10], where our generative model builds a new type of user interface with the latent space to augment human intelligence tasks, such as creating a good layout and analyzing the patterns of layout metrics.…”

Section: Usage Scenariomentioning

confidence: 99%

A Deep Generative Model for Graph Layout

Kwon

2020

IEEE Trans. Visual. Comput. Graphics

View full text Add to dashboard Cite

Fig. 1. Generative modeling of layouts for the Les Misérables character co-occurrence network [55]. We train our generative model to construct a 2D latent space by learning from a collection of example layouts (training samples). From the grid of generated samples, we can see the smooth transitions between the different layouts. This shows that our model is capable of learning and generalizing abstract concepts of graph layouts, not just memorizing the training samples. Users can use this sample grid of the latent space as a WYSIWYG interface to generate a layout they want, without either blindly tweaking parameters of layout methods or requiring expert knowledge of layout methods. The color mapping of the latent space represents the shape-based metric [23] of the generated samples. Throughout the paper, unless otherwise specified, the node color represents the hierarchical community structure of the graph [68,79], so readers can easily compare node locations in different layouts. An interactive demo is available in the supplementary material [1].Abstract-Different layouts can characterize different aspects of the same graph. Finding a "good" layout of a graph is thus an important task for graph visualization. In practice, users often visualize a graph in multiple layouts by using different methods and varying parameter settings until they find a layout that best suits the purpose of the visualization. However, this trial-and-error process is often haphazard and time-consuming. To provide users with an intuitive way to navigate the layout design space, we present a technique to systematically visualize a graph in diverse layouts using deep generative models. We design an encoder-decoder architecture to learn a model from a collection of example layouts, where the encoder represents training examples in a latent space and the decoder produces layouts from the latent space. In particular, we train the model to construct a two-dimensional latent space for users to easily explore and generate various layouts. We demonstrate our approach through quantitative and qualitative evaluations of the generated layouts. The results of our evaluations show that our model is capable of learning and generalizing abstract concepts of graph layouts, not just memorizing the training examples. In summary, this paper presents a fundamentally new approach to graph visualization where a machine learning model learns to visualize a graph from examples without manually-defined heuristics.Machine learning approaches to graph-structured data, such as social networks and biological networks, require an effective representation of the graph structure. Recently, many graph neural networks (GNNs) have been proposed for representation learning on graphs, such as graph convolutional networks [53], GraphSAGE [39], and graph isomorphism networks [84]. They have achieved state-of-the-art performance for many tasks, such as graph classification, node classification, and link prediction. We also use GNNs for learning the complex relationship between ...

show abstract

“…By interacting with each other, humans and machines can potentially solve problems and achieve better outcomes than each could on their own. Such an approach, which emphasises interactions between humans and machines, can be referred to as Artificial Intelligence Augmentation or simply Intelligence Augmentation (Carter and Nielsen, 2017). 11 More figuratively, the term "centaur" has been used to describe human-AI collaborationa reference to the mythological halfhuman, half-horse creature (Case, 2018).…”

Section: Artificial Intelligence Augmentationmentioning

confidence: 99%

Hello, World

Berryhill¹,

Heang²,

Clogher³

et al. 2019

OECD Working Papers on Public Governance

View full text Add to dashboard Cite

Artificial Intelligence (AI) is an area of research and technology application that can have a significant impact on public policies and services in many ways. In just a few years, it is expected that the potential will exist to free up nearly one-third of public servants' time, allowing them to shift from mundane tasks to high-value work. Governments can also use AI to design better policies and make better decisions, improve communication and engagement with citizens and residents, and improve the speed and quality of public services. While the potential benefits of AI are significant, attaining them is not an easy task. Government use of AI trails that of the private sector; the field is complex and has a steep learning curve; and the purpose of, and context within, government are unique and present a number of challenges.

show abstract

Using Artificial Intelligence to Augment Human Intelligence

Cited by 108 publications

References 10 publications

A Learned Representation for Scalable Vector Graphics

A Learned Representation for Scalable Vector Graphics

A Deep Generative Model for Graph Layout

Hello, World

Contact Info

Product

Resources

About