Artificial Intelligence Stack Exchange Community Digest

Top new questions this week:

Relevance of Weisfeiler–Lehman Graph Isomorphism Test limitation for Graph Neural Networks

Graph Neural Networks power is limited by the power of Weisfeiler–Lehman Graph Isomorphism algorithm. Quoting wikipedia: It has been demonstrated that GNNs cannot be more expressive than the …

graph-neural-networks geometric-deep-learning graphs graph-isomorphism-network

asked by Rexcirus Score of 3 answered by Chillston Score of 1

What is the motivation for using Q-Learning in RL?

In Spinning Up by OpenAI, it says the following regarding policy optimization methods and Q-Learning as ways of getting a good policy for RL. Trade-offs Between Policy Optimization and Q-Learning. …

reinforcement-learning deep-rl q-learning

asked by Justin T Score of 1 answered by pythonic833 Score of 1

“Attention is all you need” paper : How are the Q, K, V values calculated?

The seminal Attention is all you need paper introduces Transformers and implements the attention mecanism with “queries, keys, values”, in an analogy to a retrieval system. I understand the …

transformer attention word-embedding

asked by Soltius Score of 1

In the Dropout paper, why would increasing the dropout increase the error rate if the capacity is constant?

In the original paper on dropout, in section 7.3.2, we see that while keeping $pn$ constant, we get a (test) error increase by decreasing retainment below 0.6. Why would that happen? If $pn$ is …

deep-learning training papers regularization dropout

asked by Apples14 Score of 1 answered by Mariusmarten Score of 0

Formally, what are the layers in an Artificial Neural Network?

You may not believe it, but I am an ANN expert. Perhaps, for that reason, I am unable to grasp completely what the layers are in a Deep Forward Artificial Neural Network (DFANN). According to the Deep …

neural-networks deep-learning definitions deep-neural-networks layers

asked by neoglez Score of 1 answered by NikoNyrh Score of 1

Machine learning applied to space-time diagrams of cellular automata

I wonder if machine learning has ever been applied to space-time diagrams of cellular automata. What comprises a training set seems clear: a number of space-time diagrams of one or several (elementary)…

machine-learning convolutional-neural-networks reference-request cellular-neural-networks

asked by Hans-Peter Stricker Score of 1

What does Deep Q-Learning (DQL) do?

Hello 🙂 I’m required to write a document where I describe what DQL does in short. This is what I wrote: DQL: instead of a Q-table, a DNN is used to approximate the Q-values for each action based on a …

deep-learning deep-rl dqn deep-neural-networks

asked by Ness Score of 1

Greatest hits from previous weeks:

How is BERT different from the original transformer architecture?

As far as I can tell, BERT is a type of Transformer architecture. What I do not understand is: How is Bert different from the original transformer architecture? What tasks are better suited for BERT,…

natural-language-processing comparison transformer bert

asked by chessprogrammer Score of 19 answered by nbro Score of 20

How can neural networks deal with varying input sizes?

As far as I can tell, neural networks have a fixed number of neurons in the input layer. If neural networks are used in a context like NLP, sentences or blocks of text of varying sizes are fed to a …

neural-networks machine-learning deep-learning convolutional-neural-networks natural-language-processing

asked by Asciiom Score of 89 answered by BlindKungFuMaster Score of 62

Can a neural network be used to predict the next pseudo random number?

Is it possible to feed a neural network the output from a random number generator and expect it learn the hashing (or generator) function, so that it can predict what will be the next generated pseudo-…

neural-networks machine-learning deep-learning prediction randomness

asked by AshTyson Score of 28 answered by Demento Score of 19

What is sample efficiency, and how can importance sampling be used to achieve it?

For instance, the title of this paper reads: “Sample Efficient Actor-Critic with Experience Replay”. What is sample efficiency, and how can importance sampling be used to achieve it?

reinforcement-learning papers importance-sampling sample-efficiency

asked by Gokul NC Score of 24 answered by Jaden Travnik Score of 27

Why do we need explainable AI?

If the original purpose for developing AI was to help humans in some tasks and that purpose still holds, why should we care about its explainability? For example, in deep learning, as long as the …

philosophy social explainable-ai

asked by malioboro Score of 69 answered by Djib2011 Score of 75

What is a fully convolution network?

I was surveying some literature related to Fully Convolutional Networks and came across the following phrase, A fully convolutional network is achieved by replacing the parameter-rich fully …

machine-learning convolutional-neural-networks computer-vision image-segmentation fully-convolutional-networks

asked by pyWalker2797 Score of 16 answered by nbro Score of 21

How do I handle large images when training a CNN?

Suppose that I have 10K images of sizes $2400 \times 2400$ to train a CNN. How do I handle such large image sizes without downsampling? Here are a few more specific questions. Are there any …

neural-networks deep-learning image-recognition convolutional-neural-networks

asked by WaterRocket8236 Score of 25 answered by Maxim Score of 22

Can you answer this question?

Backpropagation with multiple output neurons but only one loss value

Suppose we have the following neural network (in reality it is a CNN with 60k parameters): This image, as well as the terminology used here, is borrowed from Matt Mazur As is visible, there are two …

neural-networks backpropagation multi-label-classification

asked by Value_Investor Score of 1