Liquid Neural Networks Do More With Less

April 30, 2023 by Joseph Long 11 Comments

[Ramin Hasani] and colleague [Mathias Lechner] have been working with a new type of Artificial Neural Network called Liquid Neural Networks, and presented some of the exciting results at a recent TEDxMIT.

Liquid neural networks are inspired by biological neurons to implement algorithms that remain adaptable even after training. [Hasani] demonstrates a machine vision system that steers a car to perform lane keeping with the use of a liquid neural network. The system performs quite well using only 19 neurons, which is profoundly fewer than the typically large model intelligence systems we’ve come to expect. Furthermore, an attention map helps us visualize that the system seems to attend to particular aspects of the visual field quite similar to a human driver’s behavior.

Mathias Lechner and Ramin Hasani — [Mathias Lechner] and [Ramin Hasani]

The typical scaling law of neural networks suggests that accuracy is improved with larger models, which is to say, more neurons. Liquid neural networks may break this law to show that scale is not the whole story. A smaller model can be computed more efficiently. Also, a compact model can improve accountability since decision activity is more readily located within the network. Surprisingly though, liquid neural network performance can also improve generalization, robustness, and fairness.

A liquid neural network can implement synaptic weights using nonlinear probabilities instead of simple scalar values. The synaptic connections and response times can adapt based on sensory inputs to more flexibly react to perturbations in the natural environment.

We should probably expect to see the operational gap between biological neural networks and artificial neural networks continue to close and blur. We’ve previously presented on wetware examples of building neural networks with actual neurons and ever advancing brain-computer interfaces.

Continue reading “Liquid Neural Networks Do More With Less” →

Why LLaMa Is A Big Deal

March 22, 2023 by Matthew Carlson 26 Comments

You might have heard about LLaMa or maybe you haven’t. Either way, what’s the big deal? It’s just some AI thing. In a nutshell, LLaMa is important because it allows you to run large language models (LLM) like GPT-3 on commodity hardware. In many ways, this is a bit like Stable Diffusion, which similarly allowed normal folks to run image generation models on their own hardware with access to the underlying source code. We’ve discussed why Stable Diffusion matters and even talked about how it works.

LLaMa is a transformer language model from Facebook/Meta research, which is a collection of large models from 7 billion to 65 billion parameters trained on publicly available datasets. Their research paper showed that the 13B version outperformed GPT-3 in most benchmarks and LLama-65B is right up there with the best of them. LLaMa was unique as inference could be run on a single GPU due to some optimizations made to the transformer itself and the model being about 10x smaller. While Meta recommended that users have at least 10 GB of VRAM to run inference on the larger models, that’s a huge step from the 80 GB A100 cards that often run these models.

While this was an important step forward for the research community, it became a huge one for the hacker community when [Georgi Gerganov] rolled in. He released llama.cpp on GitHub, which runs the inference of a LLaMa model with 4-bit quantization. His code was focused on running LLaMa-7B on your Macbook, but we’ve seen versions running on smartphones and Raspberry Pis. There’s even a version written in Rust! A rough rule of thumb is anything with more than 4 GB of RAM can run LLaMa. Model weights are available through Meta with some rather strict terms, but they’ve been leaked online and can be found even in a pull request on the GitHub repo itself. Continue reading “Why LLaMa Is A Big Deal” →

Will A.I. Steal All The Code And Take All The Jobs?

March 8, 2023 by Joseph Long 58 Comments

New technology often brings with it a bit of controversy. When considering stem cell therapies, self-driving cars, genetically modified organisms, or nuclear power plants, fears and concerns come to mind as much as, if not more than, excitement and hope for a brighter tomorrow. New technologies force us to evolve perspectives and establish new policies in hopes that we can maximize the benefits and minimize the risks. Artificial Intelligence (AI) is certainly no exception. The stakes, including our very position as Earth’s apex intellect, seem exceedingly weighty. Mathematician Irving Good’s oft-quoted wisdom that the “first ultraintelligent machine is the last invention that man need make” describes a sword that cuts both ways. It is not entirely unreasonable to fear that the last invention we need to make might just be the last invention that we get to make.

Artificial Intelligence and Learning

Artificial intelligence is currently the hottest topic in technology. AI systems are being tasked to write prose, make art, chat, and generate code. Setting aside the horrifying notion of an AI programming or reprogramming itself, what does it mean for an AI to generate code? It should be obvious that an AI is not just a normal program whose code was written to spit out any and all other programs. Such a program would need to have all programs inside itself. Instead, an AI learns from being trained. How it is trained is raising some interesting questions.

Humans learn by reading, studying, and practicing. We learn by training our minds with collected input from the world around us. Similarly, AI and machine learning (ML) models learn through training. They must be provided with examples from which to learn. The examples that we provide to an AI are referred to as the data corpus of the training process. The robot Johnny 5 from “Short Circuit”, like any curious-minded student, needs input, more input, and more input.

Continue reading “Will A.I. Steal All The Code And Take All The Jobs?” →

AI Dreaming Of Time Travel

September 28, 2022 by Matthew Carlson 20 Comments

We love the intersection between art and technology, and a video made by an AI (Stable Diffusion) imagining a journey through time (Nitter) is a lovely example. The project is relatively straightforward, but as with most art projects, there were endless hours of [Xander Steenbrugge] tweaking and playing with different parts of the process until it was just how he liked it. He mentions trying thousands of different prompts and seeds — an example of one of the prompts is “a small tribal village with huts.” In the video, each prompt got 72 frames, slowly increasing in strength and then decreasing as the following prompt came along.

There are other AI videos on YouTube, often putting the lyrics of a song into AI-generated form. But if you’ve worked with AI systems, you’ll notice that the background stays remarkably stable in [Xander]’s video as it goes through dozens of feedback loops. This is difficult to do as you want to change the image’s content without changing the look. So he had to write a decent amount of code to try and maintain visual temporal cohesion over time. Hopefully, we’ll see an open-source version of some of his improvements, as he mentioned on Twitter.

In the meantime, we get to sit back and enjoy something beautiful. If you still aren’t convinced that Stable Diffusion isn’t a big deal, perhaps we can do a little more to persuade your viewpoint.

Continue reading “AI Dreaming Of Time Travel” →

AI Midjourney Imagines “Stairway To Heaven”

September 14, 2022 by Dave Rowntree 36 Comments

This modern era of GPU-accelerated AI applications have their benefits. Pulling useful information out of mountains of raw data, alerting users to driving hazards, or just keeping an eye on bee populations are all helpful. Lately there has been a rise in attempts at producing (or should that be curating?) works of art out of carefully sculpted inputs.

One such AI art project is midjourney, which can be played with via a Discord integration bot. That bot takes some textual input, then “dreams” with it, producing sometime uncanny, often downright disturbing images.

You can have a tinker with it for free, for a short while, but there is monthly cost if you want to use it ‘for real’ whatever that means. YouTuber [Daara] has been feeding the lyrics from Led Zeppelin’s “Stairway to Heaven” into it, producing a video tour of the resulting outputs for your perusal. Continue reading “AI Midjourney Imagines “Stairway To Heaven”” →

Chinese Anti-Porn Helmet Raises Eyebrows, Questions

July 14, 2022 by Kristina Panos 71 Comments

Did you know that pornography is completely illegal in China? Probably not surprising news, though, right? The country has already put measures in place to scour the Internet in search of explicit content, mostly using AI. But the government also employs human porn appraisers, called jian huang shi, whose job it is to judge images and videos to decide whether they contain explicit content. Also probably not surprising is that humans are better than AI at knowing porn when they see it — or at least, they are faster at identifying it. Weirdness and morality and everything else aside, these jian huang shi are regular people, and frankly, they get exhausted looking at this stuff all day.

So what is the answer to burnout in this particular field? Researchers at Beijing Jiaotong University have come up with a way to bring the technological and human aspects of their existing efforts together. They’ve created a helmet that can detect particular spikes in brainwaves that occur from exposure to explicit imagery. Basically, it flashes a combination of naughty and ho-hum images in rapid succession until a spike is detected, then it flags the offending image.

Continue reading “Chinese Anti-Porn Helmet Raises Eyebrows, Questions” →

A Baudot Code Speaking Chatterbot With A Freakish Twist

April 26, 2022 by Dave Rowntree 2 Comments

[Sam Battle] known on YouTube as [Look Mum No Computer] is mostly known as a musical artist, but seems lately to have taken a bit of shine to retro telecoms gear, and this latest foray is into the realm of the minicom tty device which was a lifeline for those not blessed with ability to hear well enough to communicate via telephone. Since in this modern era of chatting via the internet, it is becoming much harder to actually find another user with a minicom, [Sam] decided to take the human out of the loop entirely and have the minicom user talk instead to a Raspberry Pi running an instance of MegaHal, which is 1990s era chatterbot. The idea of this build (that became an exhibit in this museum is not obsolete) was to have an number of minicom terminals around the room connected via the internal telephone network (and the retro telephone exchange {Sam] maintains) to a line interface module, based upon the Mitel MH88422 chip. This handy device allows a Raspberry Pi to interface to the telephone line, and answer calls, with all the usual handshaking taken care of. The audio signal from the Mitel interface is fed to the Pi via a USB audio interface (since the Pi has no audio input) module.

Continue reading “A Baudot Code Speaking Chatterbot With A Freakish Twist” →