Questions tagged [neural-network]
The neural-network tag has no summary.
36 questions
0
votes
0
answers
84
views
Fourier transform of a positive function
Let $f:\mathbb{R}\rightarrow\mathbb{R}^+$ be a positive function i.e. $f(x) > 0$. Then, what can be said about it's Fourier transform ( or it's FFT )? We can even restrict $f$ to a positive, even, ...
- 898
1
vote
0
answers
61
views
Guidelines for image detection model for static sample
I have 20,000 plus images of art (paintings, sculptures, jars, etc) stored in a data base. The actual pieces are distributed in multiple warehouses. Ideally, the physical pieces SHOULD have a sticker ...
0
votes
0
answers
57
views
How can I calculate ROC50 in python?
I need to calculate ROC50 for a classifier in python. The ROC50 value is defined as the AUC when the 50th true negative is found. I have tried setting the max fpr value for roc_auc_score in sklearn to ...
- 101
5
votes
3
answers
274
views
Benchmark Neural Networks on High-Dimensional Functions
For a personal project, I am interested in benchmarking certain neural network architectures in the context of high-dimensional function approximation. Specifically, I am interested in continuous, ...
- 159
0
votes
0
answers
58
views
representing firing rates of a neuron using delta functions [computational neuroscience]
I'm reading 'fundamentals of computaional neuroscience' by Thomas P. Trappenberg and was confused while reading about representing firing rates using direc delta functions.
instantaneous firing rate ...
0
votes
2
answers
163
views
Physics informed neural network package for hyperparameter tuning
Is there any good PINN package or algorithm for hyperparameter tuning to accelerate convergence?
- 325
3
votes
3
answers
282
views
How to implement the following operation in pytorch (tensor by equating indices)
I wasn't sure if I should post this on stackoverflow rather than here, but because I have to construct a specific tensor I think here is more suitable.
I have 2 tensors, $x \in \mathbb{R}^{M \times N \...
- 163
2
votes
0
answers
81
views
Solving systems of the form $y_i=UW x_i$ for $U,W$
I'm looking for pointers/examples of solving system of equations $y_i=f_W(f_U(x_i))$ for $W,U$ where
$f_M(x) \approx M x$
$U,W$ are updated simultaneously
$i\in (0, 10^{12})$
Simplest example is ...
- 2,779
0
votes
1
answer
100
views
Neural Network for Couette Flow
I'm trying to implement a simple Neural Network for Couette Flow. I'm working with a Fully Connected Neural Network. I'm finding that the convergence of the Neural Network is highly dependent on the ...
- 898
2
votes
1
answer
149
views
"Don't take the derivative of the approximation but approximate the derivative"..or something like this
Don't take the derivative of the approximation but approximate the derivative
or something similar.
I don't quite remember where I heard this but I am trying to find some work on the support or ...
- 21
0
votes
0
answers
145
views
How expensive is it to compute an image convolution
I was wondering how computationally expensive it is to compute an image convolution. That is, to convolve an NxN image with a 3x3 or 5x5 convolution filter? It seems like this would be a costly ...
- 297
0
votes
0
answers
74
views
difference in notation of integral operator
I'm reading paper 1 and on page 4 they define the integral operator $\mathcal{K}$ as
$$
(\mathcal{K}(a;\phi)v_t)(x) := \int_D \kappa(x,y,a(x),a(y);\phi)v_t(y)dy
$$
Now in an another paper from the ...
- 898
2
votes
2
answers
2k
views
Problems solving 2D heat equation using physics-informed neural networks
I am trying to solve 2D heat equation using the physics-informed neural networks approach. The training loss is decreasing, but my final network outputs make no sense. I am using Python/Pytorch.
2D ...
1
vote
1
answer
120
views
Automatic Differentiation using foward mode on matrices
Whilst googling I see reverse mode automatic differentiation (AD) tends to be used when optimising neural networks.
Would it not be better to use forward mode and treat your input as a single variable,...
- 13
1
vote
0
answers
99
views
Linear PDE solution with constraints
Consider the following linear PDE:
$$\nabla_q V(q) - M_d(q)M^{-1}(q)\nabla_q V_d(q) = 0,$$
where $V(q)$ and $M(q)$ are known and $M_d(q)$ is a grey box function (e.x., $M_d(q)$ is fitted using a ...
- 11