Are you ready for the latest in marine mammal fitness tracking? That's right, researchers at the University of Michigan, in collaboration with marine mammal specialists at Dolphin Quest Oahu, have developed wearable sensors for marine mammals to monitor movement and behavior in order to enhance marine conservation efforts for these animals.

It's just like a smartwatch for humans, but for dolphins. These "biologging tags" are similar to the fitness trackers that we use to track our steps and calories burned, but for dolphins, they're used to estimate the energetic cost of swimming. And let me tell you, these dolphins are no slouches in the water. During the prescribed lap trials, the animals started from rest at a floating dock and swam an 80-meter lap underwater around one of the marine mammal specialists and back to the dock at speeds of up to 21 kilometers per hour.

But why do we need to track the movement and behavior of marine mammals? It's all because of human disturbances in their habitat, including climate change, overfishing, noise pollution from shipping, construction, oil exploration and navy sonar activity. These types of disturbances can interrupt important animal behaviour like foraging for fish and socialising, but measuring disturbance is difficult because the animals live under water. (read more here)

But now, with the help of these biologging tags, researchers can estimate foraging events, how many fish were consumed during a day, and connect that to estimates of how much energy dolphins use during the movement required to catch those fish. This is important for conservation because we can then use our approach to estimate energetic costs when these animals are disturbed.

So, just like we humans have our fitness trackers to help us stay on track with our health and fitness goals, these dolphins are now wearing biologging tags to help us better understand and protect them. Who knows, maybe someday we'll even have dolphin-specific workout plans.

Students Are Making Money Without Giving Up Free Time

Attention students! Want to earn extra cash without sacrificing precious time? Look no further! Krash Kourse offers a unique opportunity for you to monetize your study sessions. Simply revise and upload your study materials and start earning money with minimal effort today. Don't let your hard work go to waste, turn it into cash with Krash Kourse. Sign up now and start earning today!

🦾 Robots Learn to Speak 'Tool' with ATLA Method

Learn to Speak the Language of Power Tools and Screwdrivers

Princeton researchers have found a new way to teach robots by using human-language descriptions of tools. The new method, called Accelerated Learning of Tool Manipulation with LAnguage (ATLA), has been found to accelerate the learning of a simulated robotic arm lifting and using a variety of tools. The results build on evidence that providing richer information during artificial intelligence (AI) training can make autonomous robots more adaptive to new situations, improving their safety and effectiveness.

The team of mechanical engineers and computer scientists obtained tool descriptions by querying GPT-3, a large language model released by OpenAI in 2020 that uses a form of AI called deep learning to generate text in response to a prompt. After experimenting with various prompts, they settled on using "Describe the [feature] of [tool] in a detailed and scientific response," where the feature was the shape or purpose of the tool.

This approach is known as meta-learning, since the robot improves its ability to learn with each successive task. It's not only learning to use each tool, but also "trying to learn to understand the descriptions of each of these hundred different tools, so when it sees the 101st tool it's faster in learning to use the new tool," said Karthik Narasimhan, an assistant professor of computer science and coauthor of the study. "We're doing two things: We're teaching the robot how to use the tools, but we're also teaching it English."

The researchers measured the success of the robot in pushing, lifting, sweeping and hammering with the nine test tools, comparing the results achieved with the policies that used language in the machine learning process to those that did not use language information. In most cases, the language information offered significant advantages for the robot's ability to use new tools. (Read more here)

While the research is still in the early stages, the findings suggest that teaching robots to understand human-language descriptions of tools could be a powerful way to improve their ability to learn and adapt to new tasks. So the next time you see a robot struggling to use a tool, just remember to speak up and give it some descriptive language.