Hot answers tagged

3 votes
Accepted

Why and when do we need to normalize weights in Reinforcement Learning?

The kind of divergence that the other question experienced is a common problem with deep RL and temporal difference methods (Q-learning, SARSA, or any Actor Critic). The weight normalisation would not ...
user avatar
  • 24.5k
2 votes

Why doesn't the high precision of neural network weights improve accuracy?

First, I have not read and do not have that book. That said, I would interpret that statement in the context of the intractability of guaranteeing that the optimization function will find global ...
user avatar
2 votes

Is there any advantage in viewing weights of a neural network as random variables?

In Bayesian statistics, as opposed to frequentist statistics, you can model the parameters as random variables. Bayesian machine learning is the application of Bayesian statistics in the context of ...
user avatar
  • 34.4k
2 votes
Accepted

Not able to understand Pytorch Tensor (Weight & Biases) Size for Linear Regression

The size of the parameters tensor is depended on what type of layer that you want to build. Convolutional, fully connected, attention or even custom layer, each layer has a difference in the way it ...
user avatar
  • 832
2 votes

Why are the initial weights of neural networks randomly initialised?

It is okay to initialize the weights to zero for a simple logistic regression, but for a neural network to initialize the weights to parameters to all zero and then apply gradient descent, it won't ...
user avatar
1 vote
Accepted

Why do neural network weights have to be between 0 and 1?

Having the weights between 0 and 1 helps accelerate learning. They do not have to be between 0 and 1. Typically the weights get normalized to [-1, 1]. But it also depends on your problem.
user avatar
  • 36
1 vote
Accepted

In TD(0) with linear function approximation, why is the gradient of $\hat v(S^{\prime}, \mathbf w)$ wrt parameters $\mathbf w$ not considered?

This is formally known as a semi-gradient method. What we would like to do is to minimize $\big(v(S) - \hat v(S, w)\big)^2$, where $v(S)$ is the true value function. This would give the gradient ...
user avatar
  • 931
1 vote

model and trained model parameters on CIFAR-10

You can find pretrained models on CIFAR-10 in this GitHub repository. Also, for fun you can take any backbone trained on ImageNet from TorchVision models. Just replace the classification part from the ...
user avatar
1 vote

Correctly input additional values into CNN

Paper 1- If I'm understanding the paper correctly, the "Measurements " just represents a collection of auxiliary information. It's not necessarily a single speed measurement, but any ...
user avatar
1 vote

Correctly input additional values into CNN

I would expect the dense layers to be able to detect certain speed ranges. This neuron activates for 0-10, this one for 10-20, this one for 20-30, this one for 20-50, this one for 47.6-89.2... Of ...
user avatar

Only top scored, non community-wiki answers of a minimum length are eligible