Issue #274 - The ML Engineer ??
Alejandro Saucedo
Tech Executive @ Zalando | Chair/Advisor @ UN, ACM, LF, etc | Join 60k+ ML Newsletter
Thank you for being part of over 60,000+ ML professionals and enthusiasts who receive weekly articles & tutorials on Machine Learning & MLOps ?? You can join the newsletter for free at https://ethical.institute/mle.html ??
If you like the content please support the newsletter by sharing with your friends via ?? Email, ?? Twitter , ?? Linkedin and ?? Facebook ! If you've come across this you can join the newsletter?asdf ?for free at https://ethical.institute/mle.html ? ?
If you are a Machine Learning Practitioner looking for an interesting opportunity, I'm currently hiring for a few roles including Data Engineer Machine Learning & Data Engineer Data Platform - do check it out and do feel free to share broadly!
This week in Machine Learning:
Meta processes hundreds of trillions of AI model data inferences per day - they have announced how two 24,576-GPU clusters are taking their infra to the next level: Meta is advancing its AI infrastructure built on Grand Teton, OpenRack, and PyTorch to support the development of advanced AI models like Llama 3. Meta's AI infra enhancements cover novel network solutions, optimized storage with a focus on efficiency and scale, and significant performance optimizations for handling large-scale AI workloads. Meta's roadmap includes expanding its AI infrastructure to feature 350,000 NVIDIA H100 GPUs by the end of 2024, making it clear that the GPU-wars are at full force across all tech giants.
MIT Introduces a brand-new on Foundation Models & Generative AI in 2024: This is quite a comprehensive introduction and deep dive into the cutting edge ecosystem of foundation models and generative AI. In this course they cover a broad range of topics, including ChatGPT, Stable-Diffusion & Dall-E, Neural Networks, Supervised Learning, Representation & Unsupervised Learning, Reinforcement Learning, Generative AI, Self-Supervised Learning, Foundation Models, GANs (adversarial), Contrastive Learning, Auto-encoders, Denoising & Diffusion.
Amazon enters the race with their new large-scale zero-shot foundation model for time-series forecasting: Amazon introduces Chronos, a transformer-based architecture model often used for language models, leveraged for probabilistic time series modeling. One of the key innovations introduced is how they approach tokenization of time series data into a fixed vocabulary, which is done through scaling and quantization. Chronos leverages pretrained models, claiming to outperform traditional and deep learning methods in both known and unseen datasets. The paper also highlights future directions including fine-tuning for improved performance and extending the model's application beyond univariate forecasting, suggesting a promising avenue for leveraging developments in language modeling to address complex time series forecasting challenges. It will be interesting to see how the battle of the titans evolves with other giants such as Google and Salesforce launching their models (and how they hold their ground compared to traditional or even common baselines).
Going Beyond Jupyter - a curated set of educational resources for best practices on engineering and system design in machine learning contexts: "Beyond Jupyter" is an excellent educational resource for production machine learning practitioners, which puts together best practices for software design principles within the ML domain. This resource covers practical examples with a case study on refactoring a Jupyter notebook-based project for better maintainability and efficiency, and a review of common anti-patterns.
High quality image generation becomes more attainable for daily use in your personal computer: This resource provides a simple and practical method to reduce the VRAM requirement of the PixArt-α diffusion model to under 8GB, making it accessible on mid-range GPUs like the NVIDIA RTX 2070 or M-series Macbooks. The optimization involves a different approach to loading model components, as well as a custom inference process, which allows the generation of high-quality 1024x1024 images in just 20 seconds.
领英推荐
Upcoming MLOps Events?
The MLOps ecosystem continues to grow at break-neck speeds, making it ever harder for us as practitioners to stay up to date with relevant developments. A fantsatic way to keep on-top of relevant resources is through the great community and events that the MLOps and Production ML ecosystem offers. This is the reason why we have started curating a list of upcoming events in the space, which are outlined below.
Upcoming conferences where we're speaking:
Other upcoming MLOps conferences in 2024: ?
In case you missed our talks:
Open Source MLOps Tools?
Check out the fast-growing ecosystem of production ML tools & frameworks at the github repository which has reached over 10,000 ? github stars. We are currently looking for more libraries to add - if you know of any that are not listed, please let us know or feel free to add a PR. Four featured libraries in the GPU acceleration space are outlined below.
? If you know of any open source and open community events that are not listed do give us a heads up so we can add them!
OSS: Policy & Guidelines
As AI systems become more prevalent in society, we face bigger and tougher societal challenges. We have seen a large number of resources that aim to takle these challenges in the form of AI Guidelines, Principles, Ethics Frameworks, etc, however there are so many resources it is hard to navigate. Because of this we started an Open Source initiative that aims to map the ecosystem to make it simpler to navigate. You can find multiple principles in the repo - some examples include the following:
If you know of any guidelines that are not in the "Awesome AI Guidelines" list, please do give us a heads up or feel free to add a pull request
About us
The Institute for Ethical AI & Machine Learning is a European research centre that carries out world-class research into responsible machine learning.