Residents in cities typically use third-party platforms such as Google Maps for route planning services. While providing near real-time processing, these state of the art centralized deployments are limited to multiprocessing environments in data centers. This raises privacy concerns, increases risk for critical data and causes vulnerability to network failure. In this paper, we propose to use decentralized road side units (RSU) (owned by the city) to perform route planning. We divide the city road network into grids, each assigned an RSU where traffic data is kept locally, increasing security and resiliency such that the system can perform even if some RSUs fail. Route generation is done in two steps. First, an optimal grid sequence is generated, prioritizing shortest path calculation accuracy but not RSU load. Second, we assign route planning tasks to the grids in the sequence. Keeping in mind RSU load and constraints, tasks can be allocated and executed in any non-optimal grid but with lower accuracy. We evaluate this system using Metropolitan Nashville road traffic data. We divided the area into 613 grids, configuring load and neighborhood sizes to meet delay constraints while maximizing model accuracy. The results show that there is a 30% decrease in processing time with a decrease in model accuracy of 99% to 92.3%, by simply increasing the search area to the optimal grid's immediate neighborhood.
more »
« less
ProgRPGAN: Progressive GAN for Route Planning
Learning to route has received significant research momentum as a
new approach for the route planning problem in intelligent transportation
systems. By exploring global knowledge of geographical
areas and topological structures of road networks to facilitate route
planning, in this work, we propose a novel Generative Adversarial
Network (GAN) framework, namely Progressive Route Planning
GAN (ProgRPGAN), for route planning in road networks. The novelty
of ProgRPGAN lies in the following aspects: 1) we propose to
plan a route with levels of increasing map resolution, starting on a
low-resolution grid map, gradually refining it on higher-resolution
grid maps, and eventually on the road network in order to progressively
generate various realistic paths; 2) we propose to transfer
parameters of the previous-level generator and discriminator to the
subsequent generator and discriminator for parameter initialization
in order to improve the efficiency and stability in model learning;
and 3) we propose to pre-train embeddings of grid cells in grid maps
and intersections in the road network by capturing the network
topology and external factors to facilitate effective model learning.
Empirical result shows that ProgRPGAN soundly outperforms
the state-of-the-art learning to route methods, especially for long
routes, by 9.46% to 13.02% in F1-measure on multiple large-scale
real-world datasets. ProgRPGAN, moreover, effectively generates
various realistic routes for the same query.
more »
« less
- Award ID(s):
- 1717084
- NSF-PAR ID:
- 10303746
- Date Published:
- Journal Name:
- Proceedings of the 27th ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD 2021)
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
null (Ed.)A broad class of unsupervised deep learning methods such as Generative Adversarial Networks (GANs) involve training of overparameterized models where the number of parameters of the model exceeds a certain threshold. Indeed, most successful GANs used in practice are trained using overparameterized generator and discriminator networks, both in terms of depth and width. A large body of work in supervised learning have shown the importance of model overparameterization in the convergence of the gradient descent (GD) to globally optimal solutions. In contrast, the unsupervised setting and GANs in particular involve non-convex concave mini-max optimization problems that are often trained using Gradient Descent/Ascent (GDA). The role and benefits of model overparameterization in the convergence of GDA to a global saddle point in non-convex concave problems is far less understood. In this work, we present a comprehensive analysis of the importance of model overparameterization in GANs both theoretically and empirically. We theoretically show that in an overparameterized GAN model with a 1-layer neural network generator and a linear discriminator, GDA converges to a global saddle point of the underlying non-convex concave min-max problem. To the best of our knowledge, this is the first result for global convergence of GDA in such settings. Our theory is based on a more general result that holds for a broader class of nonlinear generators and discriminators that obey certain assumptions (including deeper generators and random feature discriminators). Our theory utilizes and builds upon a novel connection with the convergence analysis of linear timevarying dynamical systems which may have broader implications for understanding the convergence behavior of GDA for non-convex concave problems involving overparameterized models. We also empirically study the role of model overparameterization in GANs using several large-scale experiments on CIFAR-10 and Celeb-A datasets. Our experiments show that overparameterization improves the quality of generated samples across various model architectures and datasets. Remarkably, we observe that overparameterization leads to faster and more stable convergence behavior of GDA across the board.more » « less
-
Traditionally, a high-performance microscope with a large numerical aperture is required to acquire high-resolution images. However, the images’ size is typically tremendous. Therefore, they are not conveniently managed and transferred across a computer network or stored in a limited computer storage system. As a result, image compression is commonly used to reduce image size resulting in poor image resolution. Here, we demonstrate custom convolution neural networks (CNNs) for both super-resolution image enhancement from low-resolution images and characterization of both cells and nuclei from hematoxylin and eosin (H&E) stained breast cancer histopathological images by using a combination of generator and discriminator networks so-called super-resolution generative adversarial network-based on aggregated residual transformation (SRGAN-ResNeXt) to facilitate cancer diagnosis in low resource settings. The results provide high enhancement in image quality where the peak signal-to-noise ratio and structural similarity of our network results are over 30 dB and 0.93, respectively. The derived performance is superior to the results obtained from both the bicubic interpolation and the well-known SRGAN deep-learning methods. In addition, another custom CNN is used to perform image segmentation from the generated high-resolution breast cancer images derived with our model with an average Intersection over Union of 0.869 and an average dice similarity coefficient of 0.893 for the H&E image segmentation results. Finally, we propose the jointly trained SRGAN-ResNeXt and Inception U-net Models, which applied the weights from the individually trained SRGAN-ResNeXt and inception U-net models as the pre-trained weights for transfer learning. The jointly trained model’s results are progressively improved and promising. We anticipate these custom CNNs can help resolve the inaccessibility of advanced microscopes or whole slide imaging (WSI) systems to acquire high-resolution images from low-performance microscopes located in remote-constraint settings.more » « less
-
Full-chip thermal map estimation for multi-core commercial CPUs with generative adversarial learningIn this paper, we propose a novel transient full-chip thermal map estimation method for multi-core commercial CPU based on the data-driven generative adversarial learning method. We treat the thermal modeling problem as an image-generation problem using the generative neural networks. In stead of using traditional functional unit powers as input, the new models are directly based on the measurable real-time high level chip utilizations and thermal sensor information of commercial chips without any assumption of additional physical sensors requirement. The resulting thermal map estimation method, called {\it ThermGAN} can provide tool-accurate full-chip {\it transient} thermal maps from the given performance monitor traces of commercial off-the-shelf multi-core processors. In our work, both generator and discriminator are composed of simple convolutional layers with Wasserstein distance as loss function. ThermGAN can provide the transient and real-time thermal map without using any historical data for training and inferences, which is contrast with a recent RNN-based thermal map estimation method in which historical data is needed. Experimental results show the trained model is very accurate in thermal estimation with an average RMSE of 0.47C, namely, 0.63\% of the full-scale error. Our data further show that the speed of the model is faster than 7.5ms per inference, which is two orders of magnitude faster than the traditional finite element based thermal analysis. Furthermore, the new method is about 4x more accurate than recently proposed LSTM-based thermal map estimation method and has faster inference speed. It also achieves about 2x accuracy with much less computational cost than a state-of-the-art pre-silicon based estimation method.more » « less
-
more » « less
A planned special event (PSE), such as a sports game or a concert, can greatly affect the normal operations of a transportation system. To facilitate traffic, the road network is usually reconfigured, which could include road closures, reversed lanes, and limited access to parking facilities. For recurring PSEs, event-goers are often provided with recommended routes to designated parking areas in advance. Such network reconfiguration and route and parking recommendations are, however, often ad hoc in practice. This paper focuses on the PSE traffic planning problem. We propose to simultaneously consider parking, ridesharing, and network configuration. The problem is formulated as an optimization problem with integer decision variables. We developed a flow-based traffic simulation tool that is able to incorporate parking and lane changing (which cannot be ignored around ridesharing drop-off locations) to evaluate the objective function. We also developed effective and efficient heuristic solution algorithms. The models and algorithms are tested using the real network and traffic data from Super Bowl XLIX in 2015. The results show that our methods and approaches are able to produce an effective comprehensive traffic plan with reasonable computation time. For the Super Bowl XLIX case study, the resulting optimal plan is able to save 39.6% of the total vehicle-hours associated with default network configurations. Sensitivity analysis has also been conducted with respect to the compliance rate of travelers following recommended routes. It is found that the resulting near-optimal PSE traffic plans are able to tolerate some uncertainty in the compliance rate.
- Free Publicly Accessible Full Text
-
Accepted Manuscript1.0
- Conference Paper:
- https://doi.org/10.1145/3447548.3467406
