Google Foobar Challenge: Level 3. Exploring binary numbers, dynamic… | by Katy Hagerty | Nov, 2023

Katy Hagerty

Exploring binary numbers, dynamic programming, and Markov chains Photo by Rajeshwar Bachu on Unsplash The Foobar Challenge is a coding challenge hosted by Google that can be completed in either Python or Java. I completed the challenge using Python. The challenge has its own server with specific, terminal-style commands. The problems are of various difficulty … Read more

Bias, Toxicity, and Jailbreaking Large Language Models (LLMs) | by Rachel Draelos, MD, PhD | Nov, 2023

Rachel Draelos, MD, PhD

A review of recent research on concerning characteristics of LLMs The featured image is derived from the Galton box video from Wikimedia Commons (Creative Commons Attribution-Share Alike 4.0 International license). CONTENT WARNING: This post contains examples of biased, toxic text generated by LLMs. This post provides a deep dive into recent research on bias, toxicity, … Read more

Can AI solve your problem?. Three simple heuristics for recognising… | by Daniel Bakkelund | Nov, 2023

Daniel Bakkelund

Three simple heuristics for recognising AI eligible project ideas Image by TheDigitalArtist on pixabay. In a product organisation aiming to build AI capabilities into their products and services, there is always the challenge of bringing the non-AI-literates onboard the AI train. While not everybody needs to be an AI expert, it is necessary to have … Read more

Knowledge Graphs, Hardware Choices, Python Workflows, and Other November Must-Reads | by TDS Editors | Nov, 2023

TDS Editors

We’re entering the final stretch of another eventful year for data and machine learning professionals. Many of you are making one last push to learn new skills, catch up with recent research, or prepare for your next career move before the holiday season really kicks in in many parts of the world. Our selection of … Read more

Strawberry Fields Forever – O’Reilly

Pause AI? – O’Reilly

Tim O’Reilly forwarded an excellent article about the OpenAI soap opera to me: Matt Levine’s “Money Stuff: Who Controls Open AI.” I’ll skip most of it, but something caught my eye. Towards the end, Levine writes about Elon Musk’s version of Nick Bostrom’s AI that decides to turn the world to paper-clips: [Elon] Musk gave … Read more

Avoid Overfitting in Neural Networks: a Deep Dive | by Riccardo Andreoni | Nov, 2023

Riccardo Andreoni

Learn how to implement regularization techniques to boost performances and prevent Neural Network overfitting Image source: When training a deep neural network, it’s often troublesome to achieve the same performances on both the training and validation sets. A considerably higher error on the validation set is a clear flag for overfitting: the network has … Read more

Combine Multiple LoRA Adapters for Llama 2

Benjamin Marie

Add skills to your LLM without fine-tuning new adapters Image by the author — Made with an image from Pixabay Fully fine-tuning a pre-trained large language model (LLM) for different tasks is very costly. Instead, we can freeze the parameters of the LLM while only fine-tuning a few million trainable parameters added through a LoRA … Read more