9 votes

How does noise affect generalization?

We typically think of machine learning models as modeling two different parts of the training data--the underlying generalizable truth (the signal), and the randomness specific to that dataset (the ...
user avatar
9 votes
Accepted

How does noise affect generalization?

Noise in the data, to a reasonable amount, may help the network to generalize better. Sometimes, it has the opposite effect. It partly depends on the kind of noise ("true" vs. artificial). The AI FAQ ...
user avatar
7 votes
Accepted

Is pooling a kind of dropout?

Dropout and Max-pooling are performed for different reasons. Dropout is a regularization technique, which affects only the training process (during evaluation, it is not active). The goal of dropout ...
user avatar
  • 426
6 votes
Accepted

How can the generalization error be estimated?

Generalization error is the error obtained by applying a model to data it has not seen before. So, if you want to measure generalization error, you need to remove a subset from your data and don't ...
user avatar
  • 2,029
5 votes
Accepted

What are the state-of-the-art results on the generalization ability of deep learning methods?

Introduction The paper Generalization in Deep Learning provides a good overview (in section 2) of several results regarding the concept of generalisation in deep learning. I will try to describe one ...
user avatar
  • 34.3k
4 votes
Accepted

How can we get a differentiable neural network to count things?

Estimating from an observation is a function, but "really counting" is a process. Feed-forward neural networks can learn arbitrary functions from training examples, but they cannot represent ...
user avatar
  • 24.5k
4 votes

What are the techniques for detecting and preventing overfitting?

Usually you keep track of training loss and validation loss and apply proper regularization technique (such as L1, L2, dropout, DropConnect, etc.). The more interesting technique is to observe your ...
user avatar
4 votes
Accepted

Is there any research on models that make predictions by also taking into account the previous predictions?

What you're describing is called a recurrent neural network. There are a large number of designs in this family that all have the ability to remember recent inputs and use them in the processing of ...
user avatar
3 votes
Accepted

Why is the validation loss less than the training loss, and what can be said about the effect of the learning rate?

This is very difficult to tell with the information provided, but the phenomenon is something that I have encountered many times before. Sometimes this is not a bad thing, here are some possible ...
user avatar
  • 218
3 votes
Accepted

How does replacing states with latent representations help RL agents?

In short, it is much easier for the agent to learn from a smaller dimensional state space. This is because the agent must also do representation learning; i.e. it must also infer what the state is ...
user avatar
3 votes
Accepted

How can my Q-learning agent trained to solve a specific maze generalize to other mazes?

I'm going to assume here that you're using the standard, basic, simple variant of $Q$-learning that can be described as tabular $Q$-learning, where all of your state-action pairs for which you're ...
user avatar
  • 9,419
2 votes

Is pooling a kind of dropout?

I think we would consider regularization and downsampling better in this way: dropout it puts some input value (neuron) for the next layer as 0, which makes the current layer a sparse one. So it ...
user avatar
  • 121
2 votes
Accepted

What does "the expectation is taken across different possible inputs, drawn from the distribution of inputs we expect the system to encounter" mean?

The language used here is confusing me, because it is discussing a "distribution", as in a "probability distribution", but then refers to inputs, which are data gathered from outside of any ...
user avatar
  • 9,419
1 vote

How general is generalization?

I think there's a crucial point missed in the question, touched by jros answer but without further elaboration. If you train a model on domain A: single lightning condition and test it on domain B: ...
user avatar
1 vote

Is the inductive bias always a useful bias for generalisation?

The inductive bias is the prior knowledge that you incorporate in the learning process that biases the learning algorithm to choose from a specific set of functions [1]. For example, if you choose the ...
user avatar
  • 34.3k
1 vote

Is the inductive bias always a useful bias for generalisation?

Is it true that a bias is said to be inductive iff it is useful in generalising the data? Or does inductive bias can also refer to the assumptions that may cause a decrease in performance? Tom M. ...
user avatar
1 vote
Accepted

Is there a notion of generalization in unsupervised learning?

In the paper Generalization in Unsupervised Learning (2015), Abou-Moustafa and Schuurmans develop an approach to assess the generalization of an unsupervised learning algorithm $A$ on a given dataset $...
user avatar
  • 34.3k
1 vote
Accepted

When exactly am I overfitting -- contradicting metrics

From the loss graph I would conclude, that at approx 2k steps overfitting starts, so using the model at approx 2k steps would be the best choice. But looking at the precision graph, training e.g. ...
user avatar
  • 24.5k
1 vote

Is it possible that the model is overfitting when the training and validation accuracy increase?

I'll try to answer on more general questions Is it ok that model performs better on validation, then on train? It's certainly fine if you use techniques like dropout or data augmentation and the ...
user avatar
1 vote
Accepted

Does DQN generalise to unseen states in the case of discrete state-spaces?

An environment is said to have a discrete state-space, when the number of all possible states of the environment is finite. For example, $3\times3$ Tic-tac-toe game has a discrete state-space, since ...
user avatar
  • 673
1 vote

Is there a way to ensure that my model is able to recognize an unseen example?

The comments already are giving you some good tips about how to improve what your model recognizes, but I think your question goes above that asking if there's a way to ensure that it will always ...
user avatar
  • 448
1 vote

Why does the discrepancy measure involve a supremum over the hypothesis space?

The formula $G=\mathbb{E}\left[ f(Z_{T+1}) \mid \mathbf{Z}_1^T\right] - \sum_{t=1}^Tq_t \mathbb{E}\left[ f(Z_t) \mid \mathbf{Z}_1^{t-1} \right]$ actually represents a set, for all possible values of $...
user avatar
  • 34.3k
1 vote

How is the DQN able to generalise the learning to unseen states with such a loss function?

Especially in continuous space, convergence of the value function is mainly a theoretical property. Without seeing enough of the state space, as you suggest, there's no way to ensure that your Q ...
user avatar
  • 981
1 vote

Why can neural networks generalize at all?

You've asked a question which is basically one of the most important open questions about neural networks. The answer is a huge mystery - any response to this question which immediately opens with a ...
user avatar
  • 171
1 vote

When exactly is a model considered over-parameterized?

Ok so after a little more reading, I am currently satisfy with what I found for this question. Yes, the "under-parameterized" and "over-parameterized" terms do not currently have a widely accepted ...
user avatar
  • 151
1 vote

How can the generalization error be estimated?

Error Estimation is a subject with a long history. The test-set method is only one way to estimate generalization error. Others include resubstitution, cross-validation, bootstrap, posterior-...
user avatar
1 vote

How does noise affect generalization?

PS: There is already some very good answers provided here, I will merely add to this answers in the hope that someone will find this useful: Introducing noise to a dataset can indeed have a positive ...
user avatar

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