The researchers used their framework to combine elements of two different algorithms to create a new image-classification algorithm that performed 8 percent better than current state-of-the-art approaches. There is a unifying equation that underlies many classical AI algorithms. This equation describes how these algorithms learn a specific kind of relationship between data points. Each algorithm may accomplish that in a slightly different way, but the core mathematics behind each approach is the same. The researchers identified this equation and used it to reframe popular methods and arrange them into a table, categorizing each based on the approximate relationships it learns. This table, called the periodic table of machine learning, has empty spaces that predict where algorithms should exist, but which haven’t been discovered yet. The table is like the periodic table of chemical elements, which initially contained blank squares that were later filled in by scientists. The periodic table of machine learning also has empty spaces that will be filled in as new algorithms are developed. The table gives researchers a toolkit to design new algorithms without the need to rediscover ideas from prior approaches. This is according to Shaden Alshammari, an MIT graduate student and lead author of a paper on this new framework. “It’s not just a metaphor,” Alshammari says. “We’re starting to see machine learning as a system with structure that is a space we can explore rather than just guess our way through.”
The researchers used their framework to combine elements of two different algorithms to create a new image-classification algorithm that performed 8 percent better than current state-of-the-art approaches. The researchers joined forces with John Hershey, a researcher at Google AI Perception; Axel Feldmann, an MIT graduate student; William Freeman, the Thomas and Gerd Perkins Professor of Electrical Engineering and Computer Science and a member of the Computer Science and Artificial Intelligence Laboratory (CSAIL); and senior author Mark Hamilton, an MIT graduate student and senior engineering manager at Microsoft. They didn’t set out to create a periodic table of machine learning. Alshammari began studying clustering, a machine-learning technique that classifies images by learning to organize similar images into nearby clusters. She realized the clustering algorithm she was studying was similar to another classical machine-learning algorithm, called contrastive learning, and began digging deeper into the mathematics. Alshammari found that these two disparate algorithms could be reframed using the same underlying equation. “We almost got to this unifying equation by accident. Once Shaden discovered that it connects two methods, we just started dreaming up new methods to bring into this framework. Almost every single one we tried could be added in,” Hamilton says. The framework they created, information contrastive learning (I-Con), shows how a variety of algorithms can be viewed through the lens of this unifying equation. It includes everything from classification algorithms that can detect spam to the deep learning algorithms that power LLMs. The equation describes how such algorithms find connections between real data points and then approximate those connections internally. Each algorithm aims to minimize the amount of deviation between the connections it learns to approximate and the real connections in its training data.
news is a contributor at AlgoHay. We are committed to providing well-researched, accurate, and valuable content to our readers.
You May Also Like
Search Algorithms in Information Retrieval
Decoding the Power of Search Algorithms in Modern Computing In the ever-evolving landscape of computer science, search algorithms stand out...
Graph Algorithms Topological Sorting
The Fundamentals of Graph Theory A graph consists of two primary components: nodes representing discrete entities and edges signifying relationships...
Visual Algorithm Tutorials and Animations
Mastering Algorithms Through Visual Learning and Interactive Tutorials In today's fast-paced world of software development, mastering algorithms is crucial for...
Data Structures and Algorithms Together
The Building Blocks of Efficient Programming: Data Structures in Action In the world of algorithms and software development, data structures...
Unlocking Human Potential: Balancing AI Algorithms with Consumer Freedom
Addressing Biases in AI Systems: How Interview Kickstart's Flagship Machine Learning Course Can Help
