Daniel B.
  • Member for 1 year, 7 months
  • Last seen more than a month ago
Should the training data be the same in each epoch?
5 votes

Let's quickly get out our copies of Deep Learning by Goodfellow et al. (2016). More specifically, I'm referring to page 276. On this page, the authors argue for a relatively small minibatch size, ...

View answer
Is some kind of dropout used in the human brain?
5 votes

The human brain works by having neurons constantly fire at different rates. So, if the firing rate increases, the neuron is transmitting overly exciting or calming information to further neurons ...

View answer
Is CNN capable of extracting the descriptive statistics features
Accepted answer
3 votes

CNNs learn convolutional filters that get trained on finding local, recurring patterns in some kind of image/volume data. 1D convolution is actually a thing, but I think what would be more suitable ...

View answer
Why does L1 regularization yield sparse features?
Accepted answer
3 votes

In L1 regularization, the penalty term you compute for every parameter is a function of the absolute value of a given weight (times some regularization factor). Thus, irrespective of whether a weight ...

View answer
Is the size of a neural network directly linked with an increase in its inteligence?
Accepted answer
2 votes

First of all, there is no real 'intelligence' innate to artificial Neural Networks (NNs). All they do is trying to approximate a mathematical function with a certain degree of generalization (...

View answer
How to build a test set for a model in industry?
1 votes

I am not sure whether that solves your problem at hand, but one approach you could look into is k-fold Cross Validation (CV). In this approach, you split your combined train, development, and test ...

View answer
How to predict the best from a set of messages - best practice
1 votes

One way you can definitely approach the problem is by using (Deep) Reinforcement Learning (DRL). YouTube is actually using DRL as well to suggest videos to users in order to maximize users' engagement ...

View answer
What are the disadvantages to using a distance metric in character recognition prediction
1 votes

As I see it, the question boils down to the comparison between distance (function/metric) based Optical Character Recognition (OCR) and (for example) OCR done by means of Convolutional Neural Networks ...

View answer
Get object's orientation or angle after object detection
1 votes

I think the problem can be phrased (more generally) as a Pose Estimation Problem. That term might help in obtaining better search results when searching for relevant papers. One paper that I found on ...

View answer
How to manually draw a $k$-NN decision boundary with $k=1$ given the dataset and labels?
Accepted answer
1 votes

This is a rather involved task. What to do from a high-level theoretical perspective might be easy to see, but it's difficult putting that into code from scratch. Doing this in Python using existing ...

View answer
How to handle a changing in the Reinforcement Learning environment where there is increasing or decreasing in number of agents?
1 votes

I depends on your overall model architecture (and problem specification). As I understand it, you take the observations of all agents together and feed it into one model, a central controller, which ...

View answer
Do the order of the features ie channel matter for a 1d convolutional network?
Accepted answer
1 votes

Generally, order matters. A (trained) Neural Network (NN) is just a mathematical function trained on taking some given input and producing the corresponding output. So, if you train a certain node on ...

View answer
Should forecasting with neural networks only be treated as a supervised learning (regression) problem?
Accepted answer
1 votes

I think the choice of technique strongly depends on how fine-grained your forecast-predictions need to be. When it comes to forecasting by Reinforcement Learning (RL), one prominent example is the ...

View answer
Why does adding another network help in double DQN?
1 votes

As the authors of this paper state it: In $Q$-learning, the agent updates the value of executing an action in the current state, using the values of executing actions in a successive state. This ...

View answer
How to "forward" updated NN model to a transferred model?
Accepted answer
1 votes

I think there is no simple way to transfer knowledge changes between different models. If you take your initial model and create a new version of it which you use to learn some other task (like "...

View answer
In PyTorch, why does the sequence length need to be provided as the first dimension of the input tensor for an RNN?
1 votes

As it says in the documentation, you can simply reverse the order of dimensions by providing the argument batch_first=True when constructing the RNN. Then, the dimensionality will be: (batch, seq, ...

View answer
Can fully connected layers be used for feature detection?
1 votes

First of all, (FC) Neural Networks (NN) are universal function approximators. This means, that, in theory, there must exist some NN of appropriate size that is capable of doing what a CNN can do as ...

View answer
What is the advantage of having a stochastic classification procedure?
Accepted answer
1 votes

There are multiple potential reasons for having stochastic predictions (instead of categorical/binary). First, it often simplifies training and improves the training outcome when training a classifier ...

View answer
How to validate that my DQN hyperparameters are the optimal?
1 votes

If possible, I would try to calculate what the (theoretical) maximum throughput through the intersection is for a given time interval. If the control behavior that the DQN produces comes empirically ...

View answer
Why is non-linearity desirable in a neural network?
Accepted answer
0 votes

Consider what happens if you intend to train a linear classifier on replicating something trivial as the XOR function. If you program/train the classifier (of arbitrary size) such that it outputs XOR ...

View answer
Does L1/L2 Regularization help reach an optimum result faster?
0 votes

I am not aware of any empirical results regarding this question. But in theory, adding a regularization term shall make the learning task actually even harder, since there is suddenly a second loss ...

View answer
How can I design a reinforcement learning model for a game with multiple complex actions taken at a time?
0 votes

One very popular RL algorithm that is capable of predicting multiple action outputs concurrently is Proximal Policy Optimization. In that algorithm, one or more, say $n$, tuples of outputs, $(\mu, \...

View answer
How to do early classification of time series event with small dataset?
0 votes

One kind of system you could look into are Echo State Networks (ESNs). They are relatively cheap to train and can learn to predict output signals to an arbitrary degree of precision. All you need to ...

View answer
How to let the agent choose how to populate a state space matrix in RL (using python)
0 votes

I think this question is hinting at the problem of choosing an exploration strategy. The simplest strategy is to use the so called epsilon-greedy strategy (or $\epsilon$-greedy). This means that you ...

View answer
Do correlations matter when building neural networks?
Accepted answer
0 votes

If there is some correlation between features, that is what the network will ideally find out on its own and learn to utilize. So, in general, don't take correlated samples or features out of the ...

View answer
How to quickly change hand-drawn shapes to symmetrical polished shapes?
0 votes

From how it looks, the most reliable method to try out is using Hough transform. The Hough transform can be used to detect e.g. lines and circles in images (depending on which variant you are using; ...

View answer
Why do we explore after we have an accurate estimate of the value function?
0 votes

When you are training a system using stochastic gradient descent, your system will converge towards some local minimum. If the local minimum was a good one, we would be fine with it. However, we ...

View answer
Why does my "entropy generation" RNN do so badly?
0 votes

I think there are two problems with your network. The first one, always having very similar outputs, is the rather simple one. As it seems, your network suffers from the so-called, very common Mode ...

View answer
When to use AND and when to use Implies in first-order logic?
Accepted answer
0 votes

When considering the sentence "Some boys are intelligent", it makes sense to express it by ∃x boys(x) ∧ intelligent(x). This is because the existential quantifier makes sure to express that ...

View answer
From an implementation point of view, what are the main differences between an RNN and a CNN?
Accepted answer
0 votes

In Convolutional Neural Networks (CNNs) you have small kernels (or filters) that you slide over an input (e.g. image). The value resulting from the convolution of the filter with a subset of the image ...

View answer