It just got a little easier to create soft robots that adapt to the world around them. Rice University researchers have developed a 3D printing technique (they call it “4D”) for material that automatically changes to an alternate shape when subjected to an electric current, changes in temperature or simple stress. The team produced a liquid crystal polymer ‘ink’ with two exclusive sets of molecular links — one with the originally printed shape, and another by manipulating the material. In this case, scientists just had to heat or cool the material to flip it between a flat surface and a bumpy one, among other changes.

The challenge was to craft a polymer mix that could be printed in a catalyst bath without losing its shape, Rice said.

Typically, companies use GPUs as acceleration hardware in implementing deep learning in technology. But this is pricey — top of the line GPU platforms cost around $100,000. Rice researchers have now created a cost-saving alternative, an algorithm called sub-linear deep learning engine (SLIDE) that is able to do the same job of implementing deep learning, but without the specialized acceleration hardware.

The team then took a complex workload and fed it to both a top-line GPU using Google’s TensorFlow software, and a “44-core Xeon-class CPU” using SLIDE, and found the CPU could complete the training in just one hour, compared to three and a half hours for the GPU. (There is, to our knowledge, no such thing as a 44-core Xeon-class CPU, so it’s likely that the team is referring to a 22-core, 44-thread CPU.)

SLIDE works by taking a fundamentally different approach to deep learning. GPUs leverage such networks by studying huge amounts of data — often using millions or billions of neurons, and employing different neurons to recognize different types of information. But you don’t need to train every neuron on every case. SLIDE only picks the neurons that are relevant to the learning at hand.

According to Anshumali Shrivastava, assistant professor at Rice’s Brown School of Engineering, SLIDE also has the advantage of being data parallel. “By data parallel I mean that if I have two data instances that I want to train on, let’s say one is an image of a cat and the other of a bus, they will likely activate different neurons, and SLIDE can update, or train on these two independently,” he said. “This is much a better utilization of parallelism for CPUs.”

This did bring its own challenges, however. “The flipside, compared to GPU, is that we require a big memory,” he said. “There is a cache hierarchy in main memory, and if you’re not careful with it you can run into a problem called cache thrashing, where you get a lot of cache misses.” After the team published their initial findings, however, Intel got in touch to collaborate on the problem. “They told us they could work with us to make it train even faster, and they were right. Our results improved by about 50 percent with their help.”

SLIDE is a promising development for those involved in AI. It’s unlikely to replace GPU-based training any time soon, because it’s far easier to add multiple GPUs to one system than multiple CPUs. (The aforementioned $100,000 GPU system, for example, has eight V100s.) What SLIDE does have, though, is the potential to make AI training more accessible and more efficient.

Shrivastava says there’s much more to explore. “We’ve just scratched the surface,” he said. “There’s a lot we can still do to optimize. We have not used vectorization, for example, or built-in accelerators in the CPU, like Intel Deep Learning Boost. There are a lot of other tricks we could still use to make this even faster.” However, the key takeaway, Shrivastava says, is that SLIDE shows there are other ways to implement deep learning. “Ours may be the first algorithmic approach to beat GPU, but I hope it’s not the last.”

For decades, one of the challenges in replicating human tissues has been figuring out a way to get nutrients and oxygen into the tissue and how to remove waste. Our bodies use vascular networks to do this, but it’s been hard to recreate those in soft, artificial materials.

This new tool overcomes those challenges by printing thin layers of a liquid, pre-hydrogel solution, which becomes solid when it’s exposed to blue light. This allowed the scientists to create biocompatible gels with intricate internal architecture similar to the human body’s vascular networks.

The researchers relied on other open-source projects to create their tool — called the stereolithography apparatus for tissue engineering, or SLATE. And as a way of giving back, they’ve made SLATE open source, as well. Their findings were published in Science this week, and all of their experiment data is free and open to the public. While the researchers say we’re just beginning to understand the complex form and function of the body’s structures, they hope this will help make 3D-printed organs a viable option sooner rather than later.

The team is keenly aware of safety concerns. The three-piece folding design makes it effectively impossible to trigger the needle by accident, and there are plans for a case that would prevent the button from touching anything until the shot is necessary.

EpiWear is still very young, to the point where it’s made of 3D-printed parts. The students plan to refine it, however, including a smaller, more refined look that would be more acceptable on a night out. They’re even considering adding watch functionality so that it does more than sit on your wrist in ordinary situations. Should it become a practical reality, you might not have to feel awkward about carrying a life-saving injection with you — and you’d never have to worry about leaving it behind.