-
WorldSmith: Iterative and Expressive Prompting for World Building with a Generative AI
Authors:
Hai Dang,
Frederik Brudy,
George Fitzmaurice,
Fraser Anderson
Abstract:
Crafting a rich and unique environment is crucial for fictional world-building, but can be difficult to achieve since illustrating a world from scratch requires time and significant skill. We investigate the use of recent multi-modal image generation systems to enable users iteratively visualize and modify elements of their fictional world using a combination of text input, sketching, and region-b…
▽ More
Crafting a rich and unique environment is crucial for fictional world-building, but can be difficult to achieve since illustrating a world from scratch requires time and significant skill. We investigate the use of recent multi-modal image generation systems to enable users iteratively visualize and modify elements of their fictional world using a combination of text input, sketching, and region-based filling. WorldSmith enables novice world builders to quickly visualize a fictional world with layered edits and hierarchical compositions. Through a formative study (4 participants) and first-use study (13 participants) we demonstrate that WorldSmith offers more expressive interactions with prompt-based models. With this work, we explore how creatives can be empowered to leverage prompt-based generative AI as a tool in their creative process, beyond current "click-once" prompting UI paradigms.
△ Less
Submitted 25 August, 2023;
originally announced August 2023.
-
SimCURL: Simple Contrastive User Representation Learning from Command Sequences
Authors:
Hang Chu,
Amir Hosein Khasahmadi,
Karl D. D. Willis,
Fraser Anderson,
Yaoli Mao,
Linh Tran,
Justin Matejka,
Jo Vermeulen
Abstract:
User modeling is crucial to understanding user behavior and essential for improving user experience and personalized recommendations. When users interact with software, vast amounts of command sequences are generated through logging and analytics systems. These command sequences contain clues to the users' goals and intents. However, these data modalities are highly unstructured and unlabeled, mak…
▽ More
User modeling is crucial to understanding user behavior and essential for improving user experience and personalized recommendations. When users interact with software, vast amounts of command sequences are generated through logging and analytics systems. These command sequences contain clues to the users' goals and intents. However, these data modalities are highly unstructured and unlabeled, making it difficult for standard predictive systems to learn from. We propose SimCURL, a simple yet effective contrastive self-supervised deep learning framework that learns user representation from unlabeled command sequences. Our method introduces a user-session network architecture, as well as session dropout as a novel way of data augmentation. We train and evaluate our method on a real-world command sequence dataset of more than half a billion commands. Our method shows significant improvement over existing methods when the learned representation is transferred to downstream tasks such as experience and expertise classification.
△ Less
Submitted 29 July, 2022;
originally announced July 2022.
-
Bounded-degree plane geometric spanners in practice
Authors:
Frederick Anderson,
Anirban Ghosh,
Matthew Graham,
Lucas Mougeot,
David Wisnosky
Abstract:
The construction of bounded-degree plane geometric spanners has been a focus of interest since 2002 when Bose, Gudmundsson, and Smid proposed the first algorithm to construct such spanners. To date, eleven algorithms have been designed with various trade-offs in degree and stretch factor. We have implemented these sophisticated algorithms in C++ using the CGAL library and experimented with them us…
▽ More
The construction of bounded-degree plane geometric spanners has been a focus of interest since 2002 when Bose, Gudmundsson, and Smid proposed the first algorithm to construct such spanners. To date, eleven algorithms have been designed with various trade-offs in degree and stretch factor. We have implemented these sophisticated algorithms in C++ using the CGAL library and experimented with them using large synthetic and real-world pointsets. Our experiments have revealed their practical behavior and real-world efficacy. We share the implementations via GitHub for broader uses and future research.
We present a simple practical algorithm, named AppxStretchFactor, that can estimate stretch factors (obtains lower bounds on the exact stretch factors) of geometric spanners - a challenging problem for which no practical algorithm is known yet. In our experiments with bounded-degree plane geometric spanners, we find that AppxStretchFactor estimates stretch factors almost precisely. Further, it gives linear runtime performance in practice for the pointset distributions considered in this work, making it much faster than the naive Dijkstra-based algorithm for calculating stretch factors
△ Less
Submitted 6 May, 2022;
originally announced May 2022.
-
On reaction network implementations of neural networks
Authors:
David F. Anderson,
Badal Joshi,
Abhishek Deshpande
Abstract:
This paper is concerned with the utilization of deterministically modeled chemical reaction networks for the implementation of (feed-forward) neural networks. We develop a general mathematical framework and prove that the ordinary differential equations (ODEs) associated with certain reaction network implementations of neural networks have desirable properties including (i) existence of unique pos…
▽ More
This paper is concerned with the utilization of deterministically modeled chemical reaction networks for the implementation of (feed-forward) neural networks. We develop a general mathematical framework and prove that the ordinary differential equations (ODEs) associated with certain reaction network implementations of neural networks have desirable properties including (i) existence of unique positive fixed points that are smooth in the parameters of the model (necessary for gradient descent), and (ii) fast convergence to the fixed point regardless of initial condition (necessary for efficient implementation). We do so by first making a connection between neural networks and fixed points for systems of ODEs, and then by constructing reaction networks with the correct associated set of ODEs. We demonstrate the theory by constructing a reaction network that implements a neural network with a smoothed ReLU activation function, though we also demonstrate how to generalize the construction to allow for other activation functions (each with the desirable properties listed previously). As there are multiple types of "networks" utilized in this paper, we also give a careful introduction to both reaction networks and neural networks, in order to disambiguate the overlapping vocabulary in the two settings and to clearly highlight the role of each network's properties.
△ Less
Submitted 8 March, 2021; v1 submitted 25 October, 2020;
originally announced October 2020.
-
Automated pebble mosaic stylization of images
Authors:
Lars Doyle,
Forest Anderson,
Ehren Choy,
David Mould
Abstract:
Digital mosaics have usually used regular tiles, simulating the historical "tessellated" mosaics. In this paper, we present a method for synthesizing pebble mosaics, a historical mosaic style in which the tiles are rounded pebbles. We address both the tiling problem, where pebbles are distributed over the image plane so as to approximate the input image content, and the problem of geometry, creati…
▽ More
Digital mosaics have usually used regular tiles, simulating the historical "tessellated" mosaics. In this paper, we present a method for synthesizing pebble mosaics, a historical mosaic style in which the tiles are rounded pebbles. We address both the tiling problem, where pebbles are distributed over the image plane so as to approximate the input image content, and the problem of geometry, creating a smooth rounded shape for each pebble. We adapt SLIC, simple linear iterative clustering, to obtain elongated tiles conforming to image content, and smooth the resulting irregular shapes into shapes resembling pebble cross-sections. Then, we create an interior and exterior contour for each pebble and solve a Laplace equation over the region between them to obtain height-field geometry. The resulting pebble set approximates the input image while presenting full geometry that can be rendered and textured for a highly detailed representation of a pebble mosaic.
△ Less
Submitted 7 February, 2019;
originally announced February 2019.
-
Characterisation of speech diversity using self-organising maps
Authors:
Tom A. F. Anderson,
David M. W. Powers
Abstract:
We report investigations into speaker classification of larger quantities of unlabelled speech data using small sets of manually phonemically annotated speech. The Kohonen speech typewriter is a semi-supervised method comprised of self-organising maps (SOMs) that achieves low phoneme error rates. A SOM is a 2D array of cells that learn vector representations of the data based on neighbourhoods. In…
▽ More
We report investigations into speaker classification of larger quantities of unlabelled speech data using small sets of manually phonemically annotated speech. The Kohonen speech typewriter is a semi-supervised method comprised of self-organising maps (SOMs) that achieves low phoneme error rates. A SOM is a 2D array of cells that learn vector representations of the data based on neighbourhoods. In this paper, we report a method to evaluate pronunciation using multilevel SOMs with /hVd/ single syllable utterances for the study of vowels, for Australian pronunciation.
△ Less
Submitted 23 January, 2017;
originally announced February 2017.
-
An Approach to Learning Research with a Wireless Sensor Network in an Outdoor Setting
Authors:
Tom Adam Frederic Anderson,
Yean-Fu Wen
Abstract:
Automated collection of environmental data may be accomplished with wireless sensor networks (WSNs). In this paper, a general discussion of WSNs is given for the gathering of data for educational research. WSNs have the capability to enhance the scope of a researcher to include multiple streams of data: environmental, location, cyberdata, video, and RFID. The location of data stored in a databas…
▽ More
Automated collection of environmental data may be accomplished with wireless sensor networks (WSNs). In this paper, a general discussion of WSNs is given for the gathering of data for educational research. WSNs have the capability to enhance the scope of a researcher to include multiple streams of data: environmental, location, cyberdata, video, and RFID. The location of data stored in a database can allow reconstruction of the learning activity for the evaluation of significance at a later time. A brief overview of the technology forms the basis of an exploration of a setting used for outdoor learning.
△ Less
Submitted 5 May, 2008;
originally announced May 2008.
