Issue 58 25th June 2024

Highlights

Nvidia becomes world’s most valuable company amid AI boom

Chipmaker dethrones Microsoft and Apple as stock market surge boosts valuation above $3.34tn Nvidia became the world’s most valuable company on Tuesday, overtaking tech heavyweight Microsoft, as its chips continue to play a central role in a race to dominate the market for artificial intelligence.

DeepMind researchers realize AI is really, really unfunny. That's a problem.

Big Tech companies want AI to be funny so that people use it. But 20 comedians said generative AI systems just couldn't write good jokes.

Llama Is Open-Source, But Why?

If Meta open-sources its models, wouldn’t people just build their own services instead of paying for the service (e.g., the chatbot on Meta AI, an API based on Llama, or helping you fine-tune the model and serve it efficiently) provided by Meta? With open-source models, Meta can iterate quickly with the community by directly incorporating their newly developed methods.

Research

Paper of the week: Optimize your energy consumption by applying reinforcement learning

This week's paper outlines a new method of optimizing energy usage in a smart home using reinforcement learning. Many of you will find that optimizing your energy consumption is increasingly important. This paper could be a great source of inspiration for an excellent but challenging home project.

Video

Global AI United Kingdom - Season of AI

Join Global AI United Kingdom for an insightful event focusing on harnessing the power of Microsoft Azure for AI in the cloud. Dive into Azure AI Studio - a cutting-edge 'code first' experience for building generative AI copilots.

AI Show LIVE | Prompty Project & Solving Wordle with GPT-4

Don't miss this exciting AI Show double feature! First, Seth continues his Prompty project. Then, Jennifer showcases the visual capabilities of GPT-4, shares prompt engineering techniques, and wraps up with a live Wordle demo!

Articles

London premiere of movie with AI-generated script cancelled after backlash

Plans to show The Last Screenwriter, whose script is credited to ‘ChatGPT 4.0’, prompted complaints although the film-makers insist the feature is ‘a contribution to the cause’ A cinema in London has cancelled the world premiere of a film with a script generated by AI after a backlash.

CLIP, LLaVA, and the Brain

What stood out to me was that vision transformers, CLIP, and LLaVA perform a type of processing analogous to pre-attentive visual processing in the brain. LLaVA takes the vision transformer model ViT-L/14 trained by CLIP for image encoding (Figure 5).

Listen to the AI-generated ripoff songs that got Udio and Suno sued

Jason Koebler reports on the lawsuit filed today by the RIAA against Udio and Suno, the two leading generative music startups. The lawsuit includes examples of prompts that the record labels used to recreate famous songs that were almost certainly included in the (undisclosed) training data. Jason collected some of these together into a three minute video, and the result in pretty damning. Arguing "fair use" isn't going to be easy here.

Accelerating Neural Network Training with Semi-Structured (2:4) Sparsity

Over the past year, we’ve added support for semi-structured (2:4) sparsity into PyTorch. With just a few lines of code, we were able to show a 10% end-to-end inference speedup on segment-anything by replacing dense matrix multiplications with sparse matrix multiplications.

Linear Algebra 101 for AI/ML

You don't need to be an expert in linear algebra to get started in AI, but you do need to know the basics. This is part 1 of my Linear Algebra 101 for AI/ML series, which is my attempt to compress the 6+ months I spent learning linear algebra before I started my career in AI. With the benefit of hindsight, I know now that you don't need to spend 6+ months or even 6 weeks brushing up on linear algebra to dive into AI. Instead, you can quickly ramp up on the basics and get started coding in AI much faster. As you make progress in AI/ML, you can continue your math studies.

Anthropic claims its latest model is best-in-class

Claude 3.5 Sonnet can analyze both text and images as well as generate text, and it's Anthropic's best-performing model yet — at least on paper.

New algorithm discovers language just by watching videos

DenseAV, developed at MIT, learns to parse and understand the meaning of language just by watching videos of people talking, with potential applications in multimedia search, language learning, and robotics.

Emergence thinks it can crack the AI agent code

Yet another generative AI venture has raised a bundle of money. And, like the others before it, it’s promising the moon. Emergence, whose co-founders include Satya Nitta, the former head of global AI solutions at IBM’s research division, on Monday emerged from stealth with $97.2 million in funding from Learn Capital plus credit lines totaling.

Code

GitHub - microsoft/Phi-3CookBook

This is a Phi-3 book for getting started with Phi-3. Phi-3, a family of open AI models developed by Microsoft. Phi-3 models are the most capable and cost-effective small language models (SLMs) avai...