Most Popular
1500 questions
8
votes
1 answer
How can a neural network approximate all functions when the weights are not allowed to grow exponentially?
It has been proven in the paper "Approximation by Superpositions of a Sigmoidal Function" (by Cybenko, in 1989) that neural networks are universal function approximators. I have a related question.
Assume the neural network's input and output…
Yan King Yin
- 245
- 1
- 10
8
votes
3 answers
Does an AI exist that can write software based on a formal specification?
Does an AI exist that can automatically write software based on a formal specification of the software?
user79161
- 379
- 1
- 13
8
votes
1 answer
Why are documents kept separated when training a text classifier?
Most of the literature considers text classification as the classification of documents. When using the bag-of-words and Bayesian classification, they usually use the statistic TF-IDF, where TF normalizes the word count with the number of words per…
freesoul
- 246
- 1
- 5
8
votes
1 answer
Does it make sense to use batch normalization in deep (stacked) or sparse auto-encoders?
Does it make sense to use batch normalization in deep (stacked) or sparse auto-encoders?
I cannot find any resources for that. Is it safe to assume that, since it works for other DNNs, it will also make sense to use it and will offer benefits on…
Glrs
- 231
- 3
- 8
8
votes
3 answers
How to model inhibitory synapses in the artificial neuron?
In the brain, some synapses are stimulating and some inhibiting. In the case of artificial neural networks, ReLU erases that property, since in the brain inhibition doesn't correspond to a 0 output, but, more precisely, to a negative input.
In the…
Ziemo
- 223
- 1
- 7
8
votes
2 answers
Can LSTM neural networks be sped up by a GPU?
I am training LSTM neural networks with Keras on a small mobile GPU. The speed on the GPU is slower than on the CPU. I found some articles that say that it is hard to train LSTMs (and, in general, RNNs) on GPUs because the training cannot be…
Dieshe
- 289
- 1
- 2
- 6
8
votes
5 answers
Is the smartest robot more clever than the stupidest human?
Most humans are not good at chess. They can't write symphonies. They don't read novels. They aren't good athletes. They aren't good at logical reasoning. Most of us just get up. Go to work in a factory or farm or something. Follow simple…
zooby
- 2,260
- 1
- 14
- 22
8
votes
3 answers
How should I model all available actions of a chess game in deep Q-learning?
I just read about deep Q-learning, which is using a neural network for the value function instead of a table.
I saw the example here: Using Keras and Deep Q-Network to Play FlappyBird and he used a CNN to get the Q-value.
My confusion is on the last…
malioboro
- 2,859
- 3
- 23
- 47
8
votes
3 answers
What are the state-of-the-art approaches for detecting the most important "visual attention" area of an image?
I'm trying to detect the visual attention area in a given image and crop the image into that area. For instance, given an image of any size and a rectangle of say $L \times W$ dimension as an input, I would like to crop the image to the most…
Mary
- 993
- 6
- 13
8
votes
2 answers
What are the real world uses for SAT solvers?
Why somebody would use SAT solvers (Boolean satisfiability problem) to solve their real world problems?
Are there any examples of the real uses of this model?
kenorb
- 10,525
- 6
- 45
- 95
8
votes
2 answers
What is experience replay in laymen's terms?
I've been reading Google's DeepMind Atari paper and I'm trying to understand the concept of "experience replay". Experience replay comes up in a lot of other reinforcement learning papers (particularly, the AlphaGo paper), so I want to understand…
user491626
- 241
- 1
- 5
8
votes
2 answers
Why does a one-layer hidden network get more robust to poor initialization with growing number of hidden neurons?
In a nutshell: I want to understand why a one hidden layer neural network converges to a good minimum more reliably when a larger number of hidden neurons is used. Below a more detailed explanation of my experiment:
I am working on a simple 2D…
Chrigi
- 181
- 5
8
votes
3 answers
What makes the animal brain so special?
Whenever I read any book about neural networks or machine learning, their introductory chapter says that we haven't been able to replicate the brain's power due to its massive parallelism.
Now, in modern times, transistors have been reduced to the…
user9947
8
votes
1 answer
What is an intuitive explanation of how Google's AutoML works?
I recently read that Google has developed a new AI that anyone can upload data to and it will instantly generate models, i.e. an image recognition model based on that data.
Can someone explain to me in a detailed and intuitive manner how this AI…
Seth Simba
- 1,186
- 1
- 11
- 29
8
votes
3 answers
How can 3 same size CNN layers in different ordering output different receptive field from the input layer?
Below is a quote from CS231n:
Prefer a stack of small filter CONV to one large receptive field CONV layer. Suppose that you stack three 3x3 CONV layers on top of each other (with non-linearities in between, of course). In this arrangement, each…
Inkplay_
- 421
- 4
- 8