LLMs as Optimizers
This is a really neat idea. This new paper from Google DeepMind proposes an approach where the optimization problem is described in natural language.
An LLM is then instructed to iteratively generate new solutions based on the defined problem and previously found solutions.
It was first tested on linear regression and the traveling salesman problem. Leveraging LLMs with simple prompting match or surpass hand-designed heuristic algorithms. This shows good potential for using LLMs as optimizers.
The idea is then applied to prompt optimization that aims to maximize task accuracy on different tasks like math word problem-solving.
The first piece of the proposed meta-prompt takes in previously generated prompts along with corresponding training accuracies. The second piece includes the optimization problem description with samples obtained from a training set representing the task.
At each optimization step, the goal is to generate new prompts that increase test accuracy based on the trajectory of previously generated prompts.
The optimized prompts outperform human-designed prompts on GSM8K and Big-Bench Hard, sometimes by over 50%!
For math word problem solving, one of the most effective instructions found begins with "Take a deep breath and work on this problem step-by-step".
https://t.co/GsF8fzjevX
@queryvary#ArtificialInteligence#AI
Query Vary is a game-changing tool for developers working with large language models. Streamline prompt design and testing with its comprehensive suite and enhance reliability, efficiency, and security.
https://t.co/j0PZiNC0gU
Ever felt that #LargeLanguageModels are like tackling a beast? Or pulling hairs over prompt design? Fear not, we're turning these challenges into your superpowers. Here's how. 💪🚀 #AI#MachineLearning#DevOps
4 books if you are into AI/Machine Learning:
- Designing Machine Learning Systems
- NLP with Transformers
- Introduction to Statistical Learning
- ML with PyTorch and Scikit-Learn
Check this 👇
How are you using #ChatGPT now?
I have a hunch that we all struggle when we ask questions to GPT. How do you go around that?
Happy to hear your stories and tips!
#SoftwareEngineering#GPT#ChatGPTPlus
@jordanbpeterson Anonymity does for sure. And it being a part of the digital experience, is then consequently an enabler for the opening of this pandora box.
@HansMokeNiemann Why don’t you just shut up?
You have destroyed the game we loved.
Nobody in the history of chess had such a centipawn loss ratio. Well, technically Stockfish, Lila and you, cheater.