Our new article on the @Transport_ELS Journal of Transport Geography quantifies how much additional travel people are willing to engage in to meet with other. Joint research with @BenjaGramsch@actscape@ChikaraLab
https://t.co/J7MXFhjRq1
Tarun Rambha (IITB), in a seminar hosted by Yuki Oyama, presented a parking design method using cell-based networks + Forward Hop Inequalities. This smart combo boosts computational efficiency, NYC cases showed up to 18.6% more stalls.
https://t.co/4ZiVD699Se
Our new paper on the impact of temporary pedestrianization on road traffic in a nutshell: Human-centric network design is not necessarily a zero-sum game between pedestrians and drivers.
@actscape@ChikaraLab
https://t.co/asZNu8SowI
New paper published on TR-C. This study proposes an efficient matching mechanism for two-sided crowdshipping, considering demand/supply elasticities and task-bundling. Please check it out!
https://t.co/GBEqnrvRPj
"Second-order derivatives are impossible to use in deep learning..."
Wrong !!
Here's a blog post with all you need to know about Hessian-vector products and autodiff
https://t.co/jx6NVY2QmW
Includes a benchmark of the different techniques in PyTorch and JAX.
Yesterday we presented our work on pedestrian environments at #TRBAM2024. The key takeaway message is that human-centric design is not necessarily a zero-sum game between pedestrians and drivers.
@actscape@ChikaraLab