Paladin AI and TXT e-solutions sign MOU to develop adaptive training solutions for virtual reality and augmented reality training in aviation
Montreal, Canada — Today, Paladin AI Inc. and TXT e-solutions announce they have signed a Memorandum of Understanding to cooperate on the development of an adaptive learning solution for extended reality (XR) training. TXT, along with its subsidiary PACE, is a leader in developing XR training solutions. Their flagship product, Pacelab WEAVR, provides an industrial-scale training solution with sophisticated tools for authoring, managing, and interacting in AR/VR environments. Paladin AI has developed a novel competency-based training technology that readily interfaces with learning environments and extracts industry-relevant competency metrics about the learner.
“This partnership will take training to the next level,”…
How the cult of early rising is hurting us
For a long time now we’ve heard the refrain that the most successful people rise before dawn and achieve more in their first three hours than the rest of us do all day. Or something to that effect.
Various self-help books will claim that fixing your morning will lead to health, wealth, and happiness.
It’s a tempting idea that the only thing standing between me and maximum self-actualization is setting my alarm to 5 AM and absolutely crushing my morning routine.
I used to be a believer, until I started to hit the limits of my physiology. I used to…
It’s more than just flying cars and drones
This post is an adaptation of a talk I gave the Global Airline Training & Simulation virtual conference (Global ATS-V). I provided some background and history of the field of artificial intelligence, followed by examples of how AI is currently being applied to aviation in exciting ways.
AI is eating the world
It’s quickly becoming a cliché to say that AI is disrupting every industry on the planet. It’s been almost a year now since Marc Andreessen’s famous essay in the Wall Street Journal about why software is eating the world.
Software is eating the world. — Marc Andreessen
That was true in 2011, and…
Getting Started
7 steps (with examples) to approaching any data science problem
1. Getting started
From the outside, data science can appear to be a huge and nebulous discipline. Today’s data science experts did not attend university to get data science degrees (although many universities now offer these programs).
The first generation of professional data scientists are drawn from the disciplines of mathematics, statistics, computer science, and physics.
The “science” part of data science is the classic work of posing a question, generating hypotheses, examining the evidence, and formulating a model that explains the evidence.
These are skills that anyone can learn, and there are more resources than ever to get started.
Keeping your data science workflow in the cloud
Amazon SageMaker is a powerful, cloud-hosted Jupyter Notebook service offered by Amazon Web Services (AWS). It’s used to create, train, and deploy machine learning models, but it’s also great for doing exploratory data analysis and prototyping.
While it may not be quite as beginner-friendly as some alternatives, such as Google CoLab or Kaggle Kernels, there are some good reasons why you may want to be doing data science work within Amazon SageMaker.
Let’s discuss a few.
Private data hosted in S3
Machine learning models must be trained on data. If you’re working with private data, then special care must be taken when accessing this data…
Whether you’re starting on a fresh project, or running on a remote machine, you don’t want to waste time chasing down dependencies and installing software libraries.
This tutorial will provide one of the fastest ways to get set up from a blank slate.
I have tested this approach by completing it on a plain no-frills EC2 instance running on Amazon Web Services.
Manage your packages
Deep learning workflows in PyTorch are written in the Python programming languages, and you will find yourself needing to install many additional Python packages to get all the functionality you need as a data scientist.
To keep things…
Why becoming a multi-planet species is a logical necessity
The Fermi Paradox
Launched aboard a ULA Delta II heavy lift rocket in 2009, the Kepler space telescope began its mission of searching for other planets within our Milky Way galaxy. Its telescope was trained on a small patch of sky where it continuously monitored the brightness of 156,000 stars. Over the lifespan of its mission, it detected at least 2,662 exoplanets, with more data analysis still ongoing.
Because Kepler could only detect planets whose orbits were aligned such that they periodically passed in front of their host stars, blocking part of their light, careful statistical analysis concluded that there are more planets…
And you thought a missing semicolon was bad…
Bugs are an unavoidable reality when writing software. In the age of vacuum-tube transistors and electromechanical computers, the earliest software bug was a literal insect (a moth) that crawled into the machine and got trapped in a relay. The original account was given by computing pioneer Grace Hopper, though she was not the one to find the moth. The guilty moth was taped to the log book and software errors have been called bugs ever since.
These 7 major trends will reshape how we fly
The Covid-19 pandemic struck a major blow to travel markets, hurting airlines in particular, but air travel is expected to recover by 2023. If anything, the pandemic has accelerated certain trends, such as the development of cleaner, more efficient aircraft.
Other, larger trends are at play. The convergence of several technologies, including AI, 3D printing, and virtual reality is driving much of these changes. Behind the scenes, innovators and entrepreneurs have been imagining new ways to fly. This article is an insider’s look into what’s coming down the pipe.
How to know if you’ve picked the right model
In machine learning, when faced with a mountain of unlabeled data, a data scientist’s first impulse is to try clustering the data. Clusters give us a way of describing data, finding commonalities between data points, and catching outliers.
But without any prior knowledge, how do we know how many clusters exist within the data?
Most clustering techniques require that we choose a fixed number of clusters. An algorithm like k-means will then find the centers of these k different clusters. Sometimes a visual inspection can help.
