COSMOS Brain Inspired Machine Learning

Front

What is the most appropriate loss function for a regression task?

Back

Mean Squared Error (MSE) is the most appropriate loss function for regression tasks as it measures the average squared difference between predicted and actual values.

Front

What is a discrete event in time called in a Spiking Neural Network (SNN)?

Back

In a Spiking Neural Network, the fundamental unit of information is called a 'spike' or 'action potential'.

Front

What is the Gradient Descent Equation?

Back

Front

What RuntimeError might occur when passing a batch of images through a PyTorch model?

Back

You might encounter 'RuntimeError: mat1 and mat2 shapes cannot be multiplied' if the matrix dimensions are incompatible.

Front

What is the main benefit of using logistic regression over a perceptron?

Back

Logistic regression provides probabilistic outputs, while a perceptron does not.

Front

Define Mean Squared Error (MSE).

Back

Mean Squared Error (MSE) is a loss function that calculates the average of the squares of the errors, which is the difference between predicted and actual values.

Front

What is a spike in the context of Spiking Neural Networks?

Back

A spike is a discrete event that represents the transmission of information in Spiking Neural Networks, analogous to an action potential in biological neurons.

Front

Explain the purpose of the learning rate in Gradient Descent.

Back

The learning rate determines the size of the steps taken towards the minimum of the loss function during optimization.

Front

What does the term 'overfitting' mean in machine learning?

Back

Overfitting occurs when a model learns the training data too well, capturing noise and outliers, which negatively impacts its performance on unseen data.

Front

What is the difference between classification and regression tasks?

Back

Classification tasks predict categorical outcomes, while regression tasks predict continuous numerical values.