5. Analogizers: This tribe bases its approach on psychology and focuses on learning through analogies. They work with kernel machines and are known for applications like medical diagnosis systems.

long: $NVDA $AVGO $TSM

8/x

4. Bayesians: Rooted in statistics, this group employs probabilistic inference. They use Bayes’ theorem as a fundamental tool for machine learning tasks, such as developing spam filters.

7/x

3. Evolutionaries: This tribe draws inspiration from biological evolution and natural selection. They advocate for genetic programming approaches in machine learning.

6/x

2. Connectionists: Inspired by the human brain, this group studies networks of artificial neurons. They work on theories like back-propagation and are associated with the development of neural networks and deep learning.

5/x

I believe system 2-type cognition will come out of one of the five tribes of machine learning:

1. Symbolists: This tribe focuses on inverse deduction and uses logic and philosophy as its foundation. They promote symbolic learning and aim to derive knowledge through logical reasoning.

4/x

“The 2010s were the age of scaling, now we're back in the age of wonder and discovery once again. Everyone is looking for the next thing. Scaling the right thing matters more now than ever.” -Ilya Sutskever, 11/11/24

3/x

However, these models struggle with System 2 capabilities like:
• Logical reasoning and planning
• Conscious, deliberate problem-solving
• Adapting to novel situations

2/x