Goto

Collaborating Authors

bike


BIKED: A Dataset and Machine Learning Benchmarks for Data-Driven Bicycle Design

arXiv.org Machine Learning

In this paper, we present "BIKED," a dataset comprised of 4500 individually designed bicycle models sourced from hundreds of designers. We expect BIKED to enable a variety of data-driven design applications for bicycles and generally support the development of data-driven design methods. The dataset is comprised of a variety of design information including assembly images, component images, numerical design parameters, and class labels. In this paper, we first discuss the processing of the dataset and present the various features provided. We then illustrate the scale, variety, and structure of the data using several unsupervised clustering studies. Next, we explore a variety of data-driven applications. We provide baseline classification performance for 10 algorithms trained on differing amounts of training data. We then contrast classification performance of three deep neural networks using parametric data, image data, and a combination of the two. Using one of the trained classification models, we conduct a Shapley Additive Explanations Analysis to better understand the extent to which certain design parameters impact classification predictions. Next, we test bike reconstruction and design synthesis using two Variational Autoencoders (VAEs) trained on images and parametric data. We furthermore contrast the performance of interpolation and extrapolation tasks in the original parameter space and the latent space of a VAE. Finally, we discuss some exciting possibilities for other applications beyond the few actively explored in this paper and summarize overall strengths and weaknesses of the dataset.


Comparative evaluation of CNN architectures for Image Caption Generation

arXiv.org Artificial Intelligence

Aided by recent advances in Deep Learning, Image Caption Generation has seen tremendous progress over the last few years. Most methods use transfer learning to extract visual information, in the form of image features, with the help of pre-trained Convolutional Neural Network models followed by transformation of the visual information using a Caption Generator module to generate the output sentences. Different methods have used different Convolutional Neural Network Architectures and, to the best of our knowledge, there is no systematic study which compares the relative efficacy of different Convolutional Neural Network architectures for extracting the visual information. In this work, we have evaluated 17 different Convolutional Neural Networks on two popular Image Caption Generation frameworks: the first based on Neural Image Caption (NIC) generation model and the second based on Soft-Attention framework. We observe that model complexity of Convolutional Neural Network, as measured by number of parameters, and the accuracy of the model on Object Recognition task does not necessarily co-relate with its efficacy on feature extraction for Image Caption Generation task.


CyclingNet: Detecting cycling near misses from video streams in complex urban scenes with deep learning

arXiv.org Artificial Intelligence

Cycling is a promising sustainable mode for commuting and leisure in cities, however, the fear of getting hit or fall reduces its wide expansion as a commuting mode. In this paper, we introduce a novel method called CyclingNet for detecting cycling near misses from video streams generated by a mounted frontal camera on a bike regardless of the camera position, the conditions of the built, the visual conditions and without any restrictions on the riding behaviour. CyclingNet is a deep computer vision model based on convolutional structure embedded with self-attention bidirectional long-short term memory (LSTM) blocks that aim to understand near misses from both sequential images of scenes and their optical flows. The model is trained on scenes of both safe rides and near misses. After 42 hours of training on a single GPU, the model shows high accuracy on the training, testing and validation sets. The model is intended to be used for generating information that can draw significant conclusions regarding cycling behaviour in cities and elsewhere, which could help planners and policy-makers to better understand the requirement of safety measures when designing infrastructure or drawing policies. As for future work, the model can be pipelined with other state-of-the-art classifiers and object detectors simultaneously to understand the causality of near misses based on factors related to interactions of road-users, the built and the natural environments.


A Unified Model for the Two-stage Offline-then-Online Resource Allocation

arXiv.org Artificial Intelligence

Furthermore, upon the arrival of any online agent, we have to decide quickly and irrevocably which offline agent(s) to With the popularity of the Internet, traditional offline match it with. That is mainly due to the low "patience" of resource allocation has evolved into a new the online agents. These features--online arrivals and the form, called online resource allocation. It features real-time decision-making requirement--distinguish OMMs the online arrivals of agents in the system and the from traditional matching markets where the information of real-time decision-making requirement upon the arrival all agents is fully disclosed in advance. of each online agent. Both offline and online OMMs have received significant interest in both computer resource allocation have wide applications in science and operations research communities. There is a various real-world matching markets ranging from large body of research work who studied matching policy ridesharing to crowdsourcing. There are some design for the profit maximization in ridesharing [Ashlagi emerging applications such as rebalancing in bike et al., 2019; Lowalekar et al., 2018; Bei and Zhang, 2018; sharing and trip-vehicle dispatching in ridesharing, Zhao et al., 2019; Dickerson et al., 2018a; Li et al., 2020], which involve a two-stage resource allocation process.


Discount codes to saving -- how AI apps could leave you £6,000 better off

#artificialintelligence

ARTIFICIAL intelligence for most people conjures up images from sci-fi movies – but it can help you to save money. Websites and apps can arrange your finances for you at the click of a button, saving you more than £6,000 per year. Forget A.I. that leads to robots taking over the world – and think of new, friendlier versions such as Chip and Look After My Bills. Get an app to transfer your spare cash into a savings account and watch how it mounts up. Some banks offer this service, like Lloyds' Save The Change tool, which rounds up the amount you spend on your debit card to the nearest pound and transfers the difference to a savings account.


Investigating Human Response, Behaviour, and Preference in Joint-Task Interaction

arXiv.org Artificial Intelligence

Human interaction relies on a wide range of signals, including non-verbal cues. In order to develop effective Explainable Planning (XAIP) agents it is important that we understand the range and utility of these communication channels. Our starting point is existing results from joint task interaction and their study in cognitive science. Our intention is that these lessons can inform the design of interaction agents -- including those using planning techniques -- whose behaviour is conditioned on the user's response, including affective measures of the user (i.e., explicitly incorporating the user's affective state within the planning model). We have identified several concepts at the intersection of plan-based agent behaviour and joint task interaction and have used these to design two agents: one reactive and the other partially predictive. We have designed an experiment in order to examine human behaviour and response as they interact with these agents. In this paper we present the designed study and the key questions that are being investigated. We also present the results from an empirical analysis where we examined the behaviour of the two agents for simulated users.


Can collision-aversion make motorcycles safer?

ZDNet

My motorcycle accident was a classic, the scenario you hear about in rider safety programs and read about on forums. I was cruising down a four-lane city street with no traffic in my direction but bumper-to-bumper gridlock in the oncoming lanes. At a dogleg in the road I rounded a corner to find a car from one of those oncoming lanes turning left over the double yellow into a gas station parking lot. It was textbook, something I realized even as it was happening. My motorcycle slammed into the front fender of the turning car and I came off the bike, landing on the sidewalk 15 feet away. If not for an airbag vest I wore religiously--an inflatable powered by a CO2 cartridge and clipped to the bike's frame via a tether, which acts like a rip cord when rider and machine are parted--I'm convinced I might not be writing this.


Silicon Valley Thinks Artificial Intelligence Can Upgrade Your Workouts

#artificialintelligence

When San Francisco went into COVID-19 lockdown on March 17, the last thing 32-year-old tech entrepreneur Niket Desai had to worry about was staying fit. His regular spot, Barry's, would be closed indefinitely, but Desai had installed the Tempo Studio, an all-in-one home fitness device designed to turn 30 square feet of your living room into an artificial- intelligence-powered micro gym. Tempo is a six-foot-tall weight cabinet (weights included!) While similar devices, like Tonal, offer digital resistance training at home, Tempo is the first one to deploy 3D movement analysis, combined with machine learning and AI to improve your form and curate your workouts. Its screen streams more than 200 live and on-demand classes, from a ten-minute high-intensity workout to an hour of mobility training, while its motion sensors and AI isolate up to 25 different joints at 30 frames per second.


Remote Learning's Distractions Put Extra Pressure On Students With ADHD

NPR Technology

Despite the challenges, distance learning can work well for some students with ADHD, researchers say. Some of those who aren't around peers are finding it easier to focus. Despite the challenges, distance learning can work well for some students with ADHD, researchers say. Some of those who aren't around peers are finding it easier to focus. COVID-19 forced Keriann Wilmot's son to trade his classroom for a computer.


Towards Dynamic Urban Bike Usage Prediction for Station Network Reconfiguration

arXiv.org Artificial Intelligence

Bike sharing has become one of the major choices of transportation for residents in metropolitan cities worldwide. A station-based bike sharing system is usually operated in the way that a user picks up a bike from one station, and drops it off at another. Bike stations are, however, not static, as the bike stations are often reconfigured to accommodate changing demands or city urbanization over time. One of the key operations is to evaluate candidate locations and install new stations to expand the bike sharing station network. Conventional practices have been studied to predict existing station usage, while evaluating new stations is highly challenging due to the lack of the historical bike usage. To fill this gap, in this work we propose a novel and efficient bike station-level prediction algorithm called AtCoR, which can predict the bike usage at both existing and new stations (candidate locations during reconfiguration). In order to address the lack of historical data issues, virtual historical usage of new stations is generated according to their correlations with the surrounding existing stations, for AtCoR model initialization. We have designed novel station-centered heatmaps which characterize for each target station centered at the heatmap the trend that riders travel between it and the station's neighboring regions, enabling the model to capture the learnable features of the bike station network. The captured features are further applied to the prediction of bike usage for new stations. Our extensive experiment study on more than 23 million trips from three major bike sharing systems in US, including New York City, Chicago and Los Angeles, shows that AtCoR outperforms baselines and state-of-art models in prediction of both existing and future stations.