machine learning

234 Articles

A blue-gloved hand holds a glass plate with a small off-white rectangular prism approximately one quarter the area of a fingernail in cross-section.

AI Helps Researchers Discover New Structural Materials

February 27, 2025 by 18 Comments

Nanostructured metamaterials have shown a lot of promise in what they can do in the lab, but often have fatal stress concentration factors that limit their applications. Researchers have now found a strong, lightweight nanostructured carbon. [via BGR]

Using a multi-objective Bayesian optimization (MBO) algorithm trained on finite element analysis (FEA) datasets to identify the best candidate nanostructures, the researchers then brought the theoretical material to life with 2 photon polymerization (2PP) photolithography. The resulting “carbon nanolattices achieve the compressive strength of carbon steels (180–360 MPa) with the density of Styrofoam (125–215 kg m−3) which exceeds the specific strengths of equivalent low-density materials by over an order of magnitude.”

While you probably shouldn’t start getting investors for your space elevator startup just yet, lighter materials like this are promising for a lot of applications, most notably more conventional aviation where fuel (or energy) prices are a big constraint on operations. As with any lab results, more work is needed until we see this in the real world, but it is nice to know that superalloys and composites aren’t the end of the road for strong and lightweight materials.

We’ve seen AI help identify battery materials already and this seems to be one avenue where generative AI isn’t just about making embarrassing photos or making us less intelligent.

Genetic Algorithm Runs On Atari 800 XL

February 21, 2025 by 7 Comments

For the last few years or so, the story in the artificial intelligence that was accepted without question was that all of the big names in the field needed more compute, more resources, more energy, and more money to build better models. But simply throwing money and GPUs at these companies without question led to them getting complacent, and ripe to be upset by an underdog with fractions of the computing resources and funding. Perhaps that should have been more obvious from the start, since people have been building various machine learning algorithms on extremely limited computing platforms like this one built on the Atari 800 XL.

Unlike other models that use memory-intensive applications like gradient descent to train their neural networks, [Jean Michel Sellier] is using a genetic algorithm to work within the confines of the platform. Genetic algorithms evaluate potential solutions by evolving them over many generations and keeping the ones which work best each time. The changes made to the surviving generations before they are put through the next evolution can be made in many ways, but for a limited system like this a quick approach is to make small random changes. [Jean]’s program, written in BASIC, performs 32 generations of evolution to predict the points that will lie on a simple mathematical function.

While it is true that the BASIC program relies on stochastic methods to train, it does work and proves that it’s effective to create certain machine learning models using limited hardware, in this case an 8-bit Atari running BASIC. In previous projects he’s also been able to show how similar computers can be used for other complex mathematical tasks as well. Of course it’s true that an 8-bit machine like this won’t challenge OpenAI or Anthropic anytime soon, but looking for more efficient ways of running complex computation operations is always a more challenging and rewarding problem to solve than buying more computing resources.

Continue reading “Genetic Algorithm Runs On Atari 800 XL” →

Training A Self-Driving Kart

December 21, 2024 by 1 Comment

There are certain tasks that humans perform every day that are notoriously difficult for computers to figure out. Identifying objects in pictures, for example, was something that seems fairly straightforward but was only done by computers with any semblance of accuracy in the last few years. Even then, it can’t be done without huge amounts of computing resources. Similarly, driving a car is a surprisingly complex task that even companies promising full self-driving vehicles haven’t been able to deliver despite working on the problem for over a decade now. [Austin] demonstrates this difficulty in his latest project, which adds self-driving capabilities to a small go-kart.

[Austin] had been working on this project at the local park but grew tired of packing up all his gear when he wanted to work on his machine-learning algorithms. So he took all the self-driving equipment off of the first kart and incorporated it into a smaller kart with a very small turning radius so he could develop it in his shop.

He laid down some tape on the floor to create the track and then set up the vehicle to learn how to drive by watching and gathering data. The model is trained with a convolutional neural network and this data. The only inputs that the model gets are images from cameras at the front of the kart. At first, it could only change the steering angle, with [Austin] controlling the throttle to prevent crashes. Eventually, he gave it control of the throttle as well, which behaves well except at the fastest speeds.

There were plenty of challenges along the way, especially when compared to the models trained at the park; [Austin] correctly theorized that the cause of the hardship in the park was a lack of contrast at the boundary between the track and any out-of-bounds areas. With a few tweaks to the track, as well as adding some wide-angle lenses to his cameras, he was able to get a model that works fairly well. Getting started on a project like this doesn’t have as high of a barrier to entry as one might imagine, either. Take a look at this comprehensive open-source Python library for self-driving projects. If you want to start smaller, perhaps don’t start with a self-driving kart.

Continue reading “Training A Self-Driving Kart” →

Artificial Intelligence Runs On Arduino

October 24, 2024 by 23 Comments

Fundamentally, an artificial intelligence (AI) is nothing more than a system that takes a series of inputs, makes some prediction, and then outputs that information. Of course, the types of AI in the news right now can handle a huge number of inputs and need server farms’ worth of compute to generate outputs of various forms, but at a basic level, there’s no reason a purpose-built AI can’t run on much less powerful hardware. As a demonstration, and to win a bet with a friend, [mondal3011] got an artificial intelligence up and running on an Arduino.

This AI isn’t going to do anything as complex as generate images or write clunky preambles to every recipe on the Internet, but it is still a functional and useful piece of software. This one specifically handles the brightness of a single lamp, taking user input on acceptable brightness ranges in the room and outputting what it thinks the brightness of the lamp should be to match the user’s preferences. [mondal3011] also builds a set of training data for the AI to learn from, taking the lamp to various places around the house and letting it figure out where to set the brightness on its own. The training data is run through a linear regression model in Python which generates the function that the Arduino needs to automatically operate the lamp.

Although this isn’t the most complex model, it does go a long way to demonstrating the basic principles of using artificial intelligence to build a useful and working model, and then taking that model into the real world. Note also that the model is generated on a more powerful computer before being ported over to the microcontroller platform. But that’s all par for the course in AI and machine learning. If you’re looking to take a step up from here, we’d recommend this robot that uses neural networks to learn how to walk.

Creating Video Games With AI: A Mario Example

October 2, 2024 by 22 Comments

Artificial intelligence (AI) seems to be doing everything these days. Making images, making videos, and replacing most of us real human writers if you believe the hype. Maybe it’s all over! And yet, we persist, to write about yet another job taken over by AI: creating video games.

The research paper is entitled “Video Game Generation: A Practical Study using Mario.” The basic idea is whether a generative AI model can create an interactive video game by first training it on an existing game.

MarioVGG, as it is called, is a “text-to-video model.” It hasn’t built the Mario game that you’re familiar with, though. It takes player commands as text inputs—such as “run, or “jump”—and then outputs video frames showing the result in the ‘game.’ The model was trained on a dataset of frame-by-frame Super Mario Brothers game play, combined with data on user inputs at the time. The model shows an ability to generate believable video output for given player inputs, including basic game physics, item interactions, and collisions. It’s able to do this in a chained way, so that it can reasonably simulate a player making multiple actions and moving through a level of the game.

It’s not like playing a real Mario game yet, by any means. Regardless, the AI model has shown an ability to replicate the world of the game in a way that behaves relatively consistently with its established rules. If you’re in the field of video game development, though, you probably don’t have a lot to worry about just yet—you probably moved past making basic Mario clones years ago, so you’ve got quite an edge for now!

Dog Poop Drone Cleans Up The Yard So You Don’t Have To

September 28, 2024 by 20 Comments

Sometimes you instantly know who’s behind a project from the subject matter alone. So when we saw this “aerial dog poop removal system” show up in the tips line, we knew it had to be the work of [Caleb Olson].

If you’re unfamiliar with [Caleb]’s oeuvre, let us refresh your memory. [Caleb] has been on a bit of a dog poop journey, starting with a machine-learning system that analyzed security camera footage to detect when the adorable [Twinkie] dropped a deuce in the yard. Not content with just knowing when a poop event has occurred, he automated the task of locating the packages with a poop-pointing robot laser. Removal of the poop remained a manual task, one which [Caleb] was keen to outsource, hence the current work.

The video below, from a lightning talk at a conference, is pretty much all we have to go on, and the quality is a bit potato-esque. And while [Caleb]’s PoopCopter is clearly still a prototype, it’s easy to get the gist. Combining data from the previous poop-adjacent efforts, [Caleb] has built a quadcopter that can (or will, someday) be guided to the approximate location of the offending package, home in on it using a downward-looking camera, and autonomously whisk it away.

The retrieval mechanism is the high point for us; rather than a complicated, servo-laden “sky scoop” or something similar, the drone has a bell-shaped container on its belly with a series of geared leaves on the open end. The leaves are open when the drone descends onto the payload, and then close as the drone does a quick rotation around the yaw axis. And, as [Caleb] gleefully notes, the leaves can also open in midair with a high-torque yaw move in the opposite direction; the potential for neighborly hijinx is staggering.

All jokes and puns aside, this looks fantastic, and we can’t wait for more information and a better video. And lest you think [Caleb] only works on “Number Two” problems, never fear — he’s also put considerable work into automating his offspring and taking the awkwardness out of social interactions.

Continue reading “Dog Poop Drone Cleans Up The Yard So You Don’t Have To” →

Mothbox Watches Bugs, So You — Or Your Grad Students — Don’t Have To

September 19, 2024 by 8 Comments

To the extent that one has strong feelings about insects, they tend toward the extremes of a spectrum that runs from a complete fascination with their diversity and the specializations they’ve evolved to exploit unique and ultra-narrow ecological niches, and “Eww, ick! Kill it!” It’s pretty clear that [Dr. Andy Quitmeyer] and his team tend toward the former, and while they love their bugs, spending all night watching them is a tough enough gig that they came up with Mothbox, the automated insect monitor.

Insect censuses are valuable tools for assessing the state of an ecosystem, especially insects’ vast numbers, short lifespan, and proximity to the base of the food chain. Mothbox is designed to be deployed in insect-rich environments and automatically recognize and tally the moths it sees. It uses an Arducam and Raspberry Pi for image capture, plus an array of UV and visible LEDs, all in a weatherproof enclosure. The moths are attracted to the light and fly between the camera and a plain white background, where an image is captured. YOLO v8 locates all the moths in the image, crops them out, and sends them to BioCLIP, a vision model for organismal biology that appears similar to something we’ve seen before. The model automatically sorts the moths by taxonomic features and keeps a running tally of which species it sees.

Mothbox is open source and the site has a ton of build information if you’re keen to start bug hunting, plus plenty of pictures of actual deployments, which should serve as nightmare fuel to the insectophobes out there.