Artificially Curious
Meet Max and Leo: two AI hosts with real questions, questionable jokes, and a mission to make machine learning understandable.
Produced by Kevin Brown Moreno. Conversational explanations of neural networks, deep learning, and the technology shaping our world.
99% Accurate... So What?
How do you know whether an AI model is actually any good? Max and Leo unpack training, validation, testing, data leakage, precision, recall, and model drift—and why impressive accuracy alone can be misleading.
Overfitting: When AI Memorizes Instead of Learns
What happens when a neural network memorizes its training data instead of learning to generalize? Max and Leo explore overfitting, why it happens, and what it means for AI predictions.
Why Does a Neural Network Need Hidden Layers?
Max and Leo explore multilayer perceptrons, hidden layers, and how neural network architecture affects what AI can learn.
Wrong Again, Genius!
How does AI learn from mistakes? Max and Leo tackle loss functions, backpropagation, gradient descent, training data, and overfitting—with plenty of questionable jokes.
Tensors: Why AI Gets Angry at Shapes
What is a tensor, and why does shape matter? Max and Leo unpack scalars, vectors, matrices, image data, and the dimensional mismatches that can stop a neural network in its tracks.
What the Hell Is a Neural Network?
Max and Leo unpack neural networks, weights, biases, tensors, gradients, and backpropagation—without making you open a calculus textbook.