./episode_71.sh
Success is a poor teacher.
Hey friends! Welcome to ./engineered_mind.sh – in this newsletter we explore & discuss strategies, systems & tools that help us become better, smarter and more effective scientists, gadgeteers and thinkers.
❤️ Weekly Favourite Things
🎬 My Favourite Video
🏫 Courses – One Free Month of Skillshare to explore your creativity with classes in illustration, photography, design, productivity and more!
📚 Neural Implicit Flow: A mesh-agnostic dimensionality reduction paradigm of spatio-temporal data
Common existing paradigms for dimensionality reduction include linear methods, such as the singular value decomposition (SVD), and nonlinear methods, such as variants of convolutional autoencoders (CAE). However, these encoding techniques lack the ability to efficiently represent the complexity associated with spatio-temporal data, which often requires variable geometry, non-uniform grid resolution, adaptive meshing, and/or parametric dependencies.
To resolve these practical engineering challenges, we propose a general framework called Neural Implicit Flow (NIF) that enables a mesh-agnostic, low-rank representation of large-scale, parametric, spatial-temporal data.
💻 So, what is a physics-informed neural network?
The idea is very simple: add the known differential equations directly into the loss function when training the neural network.
This is done by sampling a set of input training locations () and passing them through the network. Next gradients of the network’s output with respect to its input are computed at these locations (which are typically analytically available for most neural networks, and can be easily computed using autodifferentiation). Finally, the residual of the underlying differential equation is computed using these gradients, and added as an extra term in the loss function.
🧠 DALL·E 2 – A new AI system that can create realistic images and art from a description in natural language
DALL·E 2 uses OpenAI's CLIP image recognition system and adds the ability for users to edit the results. They can now select and edit areas of existing images, add or remove elements, mash together two images into a single collage and generate further variations of an existing image.
👉 Research paper: https://cdn.openai.com/papers/dall-e-2.pdf
💻 Engineering Tool of the Week – PTC Mathcad
Engineering calculations are essential for product design. To achieve excellence in engineering, you need a comprehensive yet intuitive application that performs accurate calculations, enables traceability, protects intellectual property, and allows you to show your work.
Document your most critical engineering calculations in an engineering notebook with natural mathematical notation and units intelligence. Show your work using rich formatting options alongside plots, text, and images in a single, professionally formatted document.
📚 Book of the Week
The Myth of Multitasking – How “Doing It All” Gets Nothing Done
Productivity and effective time management end with multitasking. The false idea that multitasking is productive has become even more prevalent and damaging to our productivity and well-being since the first edition of The Myth of Multitasking was published in 2008. In this revised and updated second edition, author and productivity expert Dave Crenshaw provides a solution for the chaos of distraction that multitasking creates—and a way to combat the temptation to constantly switch between tasks.
🙃 Meme of the Week
Typical Fluid Mechanics Papers.
🎬 Animation of the Week
✍️ Closing Remarks
Reach out to me via Twitter or Instagram or LinkedIn, so let’s connect!
For any business related issues or collaborations, feel free to write me an email to jousef@jousefmurad.com!
See you next week and in the meantime, make sure to keep engineering your mind! 🧠
Jousef