Issue #295 - The ML Engineer ??

Issue #295 - The ML Engineer ??

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 are a Machine Learning Practitioner looking for an interesting opportunity, I'm currently hiring for an Applied Science Manager for Forecasting & Causal Inference for a Senior Applied Scientist for Forecasting - do check it out and do feel free to share broadly!


This week in Machine Learning:


Salesforce's New Foundation AI

Salesforce AI steps into the foundation LLM race with their latest model MINT-1T as the first open-source multimodal model with one trillion tokens: MINT-1T drives quite a few interesting innovations such as expanding on the previously experimented datasets by incorporating diverse data sources including HTML documents, PDFs, and ArXiv papers. This scale seems to allow for better domain coverage, particularly in scientific documents with results suggesting that models trained on MINT-1T outperform those trained on prior datasets in tasks such as captioning and visual question answering.


Building a Custom ML Platform

As organisations are looking to develop their internal machine learning experimentations platform they wrestle with the question of "build vs buy" - this article showcases how Eppo replaced Airflow with an in-house solution built using NodeJs, BullMQ, and Postgres. This article provides interesting view of Airflow's limitations and high maintenance as their workloads scale, as well as showing that alternatives like Dagster and Argo CD had similar limitations. There are interesting perspectives to developing an in-house solution however it indeed is surprising to see that organisations are still not able to build their internal e2e machine learning platform with best-of-breed open source tools and have to resort to complex custom-built platforms.


25 CompSci Papers to Read

25 Computer Science Papers every programmer and MLOps practitioner should read: In Distributed Systems and Databases: 1.1 Google File System (GFS), 1.2 Amazon Dynamo, 1.3 BigTable (Google), 1.4 Cassandra (Apache), 1.5 Google Spanner, 1.6 FoundationDB, and 1.7 Amazon Aurora, focusing on scalable, fault-tolerant storage and database systems. In Data Processing and Analysis (2): 2.1 MapReduce (Google), 2.2 Hadoop (Apache), 2.3 Flink (Apache), 2.4 Kafka (Apache), 2.5 Dapper (Google), and 2.6 Monarch (Google), which revolutionized data processing and streaming. In Complex Challenges in Distributed Systems (3): 3.1 Google Borg, 3.2 Uber Schemaless, 3.3 Google Zanzibar, 3.4 Thrift (Facebook), 3.5 Raft Consensus Algorithm, and 3.6 Time, Clocks, and Ordering of Events (1978), addressing containerization, consensus algorithms, and access control. In Groundbreaking Concepts and Architectures (4): 4.1 Attention is All You Need (Transformer), 4.2 Bitcoin White Paper, and 4.3 Go-To Statement Considered Harmful, which introduced transformative ideas in NLP, blockchain, and programming practices. Finally, in Specific Applications and Optimizations (5): 5.1 Memcached, 5.2 RocksDB (MyRocks), 5.3 Twitter's Who to Follow Service, and 5.4 Survey on Vector Databases (2021), focusing on optimizations in caching, storage, recommendations, and high-dimensional data handling.


Lessons on AI Training

A great set of lessons from the Frontlines of AI Training from leading research labs in AI: This is a great resource that highlights how AI labs are innovating in data strategies, including the use of synthetic data, advanced data curation techniques, and scalable management solutions, to overcome challenges like potential data scarcity. This is a great reminder of how high-quality data is absolutely key in developing effective AI models.


Introducing Apple Foundation AI

Apple also joins the race of foundation AI models with an advanced ~3 billion parameter AI system integrated into their devices: This is an insightful innovation in on-device / edge deep learning optimized for efficiency and tailored to everyday tasks like text editing, notifications, and image creation. Apple showcases how these models are trained using Apple's AXLearn framework, and how they were able to focus on privacy and responsible AI principles.


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.

  • Kompute - Blazing fast, lightweight and mobile phone-enabled GPU compute framework optimized for advanced? data processing usecases.
  • CuPy - An implementation of NumPy-compatible multi-dimensional array on CUDA. CuPy consists of the core multi-dimensional array class, cupy.ndarray, and many functions on it.
  • Jax - Composable transformations of Python+NumPy programs: differentiate, vectorize, JIT to GPU/TPU, and more
  • CuDF - Built based on the Apache Arrow columnar memory format, cuDF is a GPU DataFrame library for loading, joining, aggregating, filtering, and otherwise manipulating data.

? 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:

  • MLSecOps Top 10 Vulnerabilities - This is an initiative that aims to further the field of machine learning security by identifying the top 10 most common vulnerabiliites in the machine learning lifecycle as well as best practices.
  • AI & Machine Learning 8 principles for Responsible ML - The Institute for Ethical AI & Machine Learning has put together 8 principles for responsible machine learning that are to be adopted by individuals and delivery teams designing, building and operating machine learning systems.
  • An Evaluation of Guidelines - The Ethics of Ethics; A research paper that analyses multiple Ethics principles.
  • ACM's Code of Ethics and Professional Conduct - This is the code of ethics that has been put together in 1992 by the Association for Computer Machinery and updated in 2018.

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.

Check out our website

要查看或添加评论,请登录