Sharing interesting resources I came across recently.

• Book: Deep Learning: A Visual Approach
  An accessible, highly-illustrated introduction to deep learning that offers visual and conceptual explanations instead of equations.
• Book: Python for Geospatial Data Analysis
  Fundamentals of spatial data analysis in Python.
• Articles: A Gentle Introduction to Graph Neural Networks, Understanding Convolutions on Graphs
  These articles on Distill.pub explore and explain modern graph neural networks and related concepts in a very nice and visual way.
• To follow: Geoff Boeing, Fabio Miranda
• Dataset: Global Human Settlement Layer
  Open and free data and tools for assessing the human presence on the planet (degree of urbanisation, population, built-up, urban centers).
• Podcast episode: Could the Next Brooklyn Be … Las Vegas?!
  
• Tool: OSMnx
  ‍Python package that lets you download geospatial data from OpenStreetMap and model, project, visualize, and analyze real-world street networks and any other geospatial geometries.
  
• Upcoming conference: ACM Geospatial
  Program incl. papers: https://sigspatial2021.sigspatial.org/program/
• Tool: OSMnx
  ‍Python package that lets you download geospatial data from OpenStreetMap and model, project, visualize, and analyze real-world street networks and any other geospatial geometries.
  
• Project: Urban Pulse
  ‍Urban Pulse is a framework that uses computational topology techniques to capture the pulse of cities.
  
• Project: Cityseer
  Package for pedestrian-scale network-based urban analysis: network analysis, landuse accessibilities and mixed uses, statistical aggregations.
  

I had the opportunity to give a talk at this year's Spatial Data Science Conference and I decided to present selected insights about mobility (non)recovery to pre-Covid levels of normalcy.

In particular, I'm looking into the level of impact on industry (consumer profiles) and on personal mobility (mobility scales).

Slides: download slides in PDF
Talk: watch on YouTube

‍

‍

Sharing interesting resources I came across recently.

• MIT Review - Mind Issue
  This issue looks at how neuroscience is learning how brain produces the mind — and all the mysterious and essential parts of being a human that go with it.
• Physics-based Deep Learning
  Practical and comprehensive introduction of everything related to deep learning in the context of physical simulations.
• Podcast: In Machines We Trust
  "Podcast about the automation of everything."
• Data: https://opentopography.org/
  Portal with high-resolution topography data and tools.
• Slides and individual videos from the Stanford Graph Learning Workshop
  Workshop on recent advances in Graph Neural Networks.
• PyG (PyTorch Geometric) 2.0
  Library for training Graph Neural Nets based on PyTorch. New release 2.0.
  Related Colab notebooks
• Neo4j adds ability to define and run pipelines for supervised machine learning (link to update)
  
• Paper: Identifying and modeling the structural discontinuities of human interactions
  
• Upcoming conference: UbiComp
  Program incl. papers: https://schedule.ubicomp.org/