- Apr 23, 2025
- 4 min read
Hello, AI Enthusiasts.
Halfway through the week and your inbox is still judging you? Same. Let’s take a smarter break with a few quick hits from the world of AI.
MIT just gave machine learning its own version of the periodic table—and it actually helps make sense of how different algorithms are connected. It’s like chemistry class, but with more neural nets.
Meanwhile, over at Boise State, what started as a small faculty AI group turned into a full-on community affair.
Also, Google wants to help cities outsmart traffic using AI. Think smoother commutes, fewer emissions, and a whole lot less road rage.
Here's another crazy day in AI:
The secret structure of machine learning methods
Faculty-led AI group grows into campuswide movement
Google’s mobility AI aims to fix city traffic with data
Some AI tools to try out
TODAY'S FEATURED ITEM: The Periodic Table That Predicts AI's Future
Image Credit: Wowza (created with Ideogram)
Have you ever wondered what would happen if we could organize AI algorithms the same way chemical elements are arranged in the periodic table?
New research from MIT introduces a structured framework that visually connects more than 20 classical machine learning algorithms—offering what the researchers call a “periodic table of machine learning.” Led by MIT graduate student Shaden Alshammari, the team developed this framework by identifying a single unifying equation that underlies many well-known algorithms.
Featured in a recent article written by Adam Zewe for MIT News, the research outlines how this shared equation could help researchers reinterpret established approaches, create combinations that haven't yet been explored, and potentially uncover new algorithms entirely. This model, called information contrastive learning (I-Con), functions as both a visual guide and a conceptual tool for navigating the broader landscape of machine learning.
Here’s what the framework lays out:
A shared mathematical foundation links a broad set of machine learning techniques
Algorithms are categorized based on how they estimate data relationships
Visual structure highlights both established techniques and theoretical gaps
A hybrid model built using the framework outperformed standard classifiers by 8%
I-Con supports more thoughtful model development by exposing underlying patterns
The approach makes connections across algorithms that previously seemed unrelated
Rather than presenting a conclusion, this work opens up more questions. It suggests that the field of machine learning may benefit from stepping back and considering the architecture of its own knowledge. Much like the periodic table organizes the elements in chemistry, this framework encourages researchers to see the relationships among algorithms not as isolated techniques, but as part of an evolving structure.
This perspective invites curiosity. What if frameworks like I-Con could help us better understand how AI evolves? How might this change the way we teach algorithms, explore interdisciplinary applications, or approach innovation? The answers aren't fixed—but having a structured way to think about them might be a valuable place to start.
Read the full article here.
Read the paper here.
OTHER INTERESTING AI HIGHLIGHTS:
Faculty-Led AI Group Grows Into Campuswide Movement
/Boise State Newsroom
What began as a small AI interest group for business faculty at Boise State has blossomed into a thriving cross-campus and community-wide collaboration. The COBE AI Brown Bag group now welcomes educators, IT professionals, students, and even representatives from companies like Google and OpenAI. Faculty use the meetings to share practical applications, from teaching with ChatGPT to exploring AI ethics and job market trends. The group has fostered surprising interdisciplinary connections and helped build a culture of shared AI exploration on campus.
Read more here.
Google’s Mobility AI Aims to Fix City Traffic with Data
/Neha Arora, Software Engineer, and Ivan Kuznetsov, Group Product Manager, on Google Research Blog
Google Research has unveiled Mobility AI, a comprehensive initiative designed to help cities solve transportation challenges using cutting-edge AI. By combining measurement, simulation, and optimization technologies, the program equips public agencies with tools to manage traffic, reduce emissions, and improve safety. From predicting parking difficulty to calibrating full-city traffic simulations, Mobility AI builds on years of research in routing, mapping, and urban mobility. Google hopes to work with agencies, planners, and researchers to bring these innovations to the streets and shape the future of transportation.
Read more here.
SOME AI TOOLS TO TRY OUT:
Deeto – Manage and grow your client reference group with ease.
Manychat – Automate messaging across IG, WhatsApp, TikTok, and Messenger.
Genspark AI Slides – Quickly generate presentation slides with an agentic AI tool.
That’s a wrap on today’s Almost Daily craziness.
Catch us almost every day—almost! 😉
EXCITING NEWS:
The Another Crazy Day in AI newsletter is on LinkedIn!!!
Leveraging AI for Enhanced Content: As part of our commitment to exploring new technologies, we used AI to help curate and refine our newsletters. This enriches our content and keeps us at the forefront of digital innovation, ensuring you stay informed with the latest trends and developments.
