Stack Exchange Network

Stack Exchange network consists of 175 Q&A communities including Stack Overflow, the largest, most trusted online community for developers to learn, share their knowledge, and build their careers.

Visit Stack Exchange

Questions tagged [artificial-intelligence]

Artificial Intelligence (AI) is a topic in computer science that deals with the study / creation of intelligent machines. Use this tag for on-topic questions that have an AI aspect.

0
votes
0answers
9 views

Zero Sum problem with MDP formulation and the difference with minimax approach

Suppose that we have formulate a zero sum game with MDP and Ua(s) and Ub(s) are the utilities of A and B in the s state. Suppose that all rewards and utilities are calculated from the A's point of ...
0
votes
1answer
45 views

What is the difference between policy-based, on-policy, value-based, off-policy, model-free and model-based?

I'm trying to clear things out for myself, there are a lot of different categorizations within RL. Some people talk about: On-policy & Off-Policy Model-based & Model-free Model-based, Policy-...
0
votes
1answer
18 views

What are the applications of different cost functions and which one to choose? [duplicate]

I have just read about different cost functions for training neural networks. How to determine which cost function is applicable in any given situation?
0
votes
0answers
29 views

Why does this Neural Network have an accuracy drop-off on higher feature sizes

I am currently training different models (neural network, support vector machine, and XGBoost) to predict concentrations of antibiotics required to prevent growth of a bacteria from whole genome ...
0
votes
0answers
11 views

Confidence vs. Count in association rule mining: which one is better?

I am writing a program that mines association rules from a large data set. I have an array of association rules, and I have to decide which ones are more representative of the patterns I am studying. ...
1
vote
1answer
24 views

Is it possible to use alpha-beta pruning for non-zero-sum games with more than two players?

I have read somewhere that the minimax algorithm can be generalized for more than two players. Imagine that we have 3 players that each of them want to maximize its own answer. Is it possible to use ...
0
votes
1answer
24 views

Understanding Feed Forward Neural Network

My problem with FFNN is that I do not understand in which use cases this network makes sense. Does anyone have an example where this is used? Once I read on the internet that it could be used for YES/...
0
votes
1answer
40 views

What is an Hopfield network?

What is a Boolean Hopfield network model? Is it same as binary hopfield network?
0
votes
1answer
28 views

Initialization for ridge regression

What should be a better initialization for weights of ridge regression if I have to perform gradient descent. I have tried with all weights 0, all weights 1, and random initialization. In all the ...
0
votes
2answers
42 views

How to get PCA of the testing data? [duplicate]

I'd like to transform my data into pca (preprocessing data before I use data into classification model). I separate my data into data training and data testing. I used princomp in R to process pca ...
0
votes
1answer
28 views

Distribution of points in SVMs

Is that an assumption that in SVMs there is a separation of -1 to +1 for all the sample data points, i.e., if x_p is a positive label point then w.x_p + b >= 1 and if x_n is a negative label point ...
0
votes
0answers
17 views

Problem - “Datewise” (for next week in future) forecast for the “Customer numbers” expected to have alarm

Please provide some direction for the problem I am trying to solve. IoT sensors at villas generate Fire and Maintenance alamrs. I have alarm history data - Customer no., timestamp of alarm, problem ...
1
vote
0answers
33 views

Advantage and Disadvantage of genetic programming [closed]

Are genetic programming and genetic algorithm the same ? If not how are they different? what are the advantage and the disadvantages of them?
0
votes
0answers
42 views

Time shifting in time series forecasting

I am working on an ANN model for univariate time series forecasting. The step size is 1, so I try to forecast value of t+1, using value of t. Unfortunately, my forecasts have time shift problem. The ...
0
votes
2answers
116 views

Stacked shallow autoencoders vs. deep autoencoders

In LeCun et. all "Deep Learning", Chapter 14, page 506, I found the following statement: "A common strategy for training a deep autoencoder is to greedily pretrain the deep architecture by training a ...
-1
votes
1answer
26 views

Problem with the word 'machine' in the definition of machine learning by Mitchell in the book “Machine Learning” [closed]

The definition : A computer program is said to learn from experience E with respect to some task T and performance measure P, if its performance at task T, as measured by P, improves with experience ...
2
votes
1answer
49 views

Independence imply d-separation?

In the context of Bayesian networks if two random variables (i.e. nodes) are d-separated, they are independent. However, is there any example of random variables being independent but not d-separated?
1
vote
0answers
71 views

How to handle missing data in PPCA

I am implementing a paper "Probabilistic Principal Component Analysis" (PPCA) which deals with a dataset where each vector suffers from at least one missing value missing values. Generally, PPCA ...
0
votes
1answer
55 views

How to apply multi agent deep reinforcement learning to an environment with discrete action space

Do you know or have heard about any cutting edge deep reinforcement-learning algorithm which can be successfully applied for discrete action-spaces in multi-agent settings? I have been researching ...
1
vote
0answers
95 views

Proof of NOT initialization of weights to zero

Can anyone provide me the mathematical details to why we shouldn't initialize a neural network (single layer ANN) weights to zero? In multi-layer we don't initialize it to zero because of loss of ...
1
vote
1answer
95 views

What is the difference between Generative Adversarial Networks (GAN) and Generative Antagonistic System (GAS)?

What is the difference between Generative Adversarial Networks (GAN) and Generative Antagonistic System (GAS) in the neural network?
0
votes
0answers
20 views

Variance of error term is nonconstant between observations

I used XGB algorithm to train a model. The task is to train models to predict human personality based on his/her personal photo. We found some significant features when we extracted them by Pearson ...
1
vote
1answer
51 views

Trouble understanding value iteration

I have trouble understanding how the value iteration algorithm for MDP:s work. I'm trying to follow the canonical grid world example (slide 17), but I don't get the correct results. Here's my work: ...
1
vote
1answer
97 views

Mobilenet Original Paper Architecture vs Keras Implementation

1. Question: Why do original paper mobilenet architecture and keras implementation differ? Keras implementation of mobilenet's last 5 layers after AVG Pool layer: ...
4
votes
1answer
250 views

What is intuition behind high variance of Monte Carlo method? [closed]

I'm studying Reinforcement learning from lectures of David Silver, where he says that Monte Carlo method is not biased and has very high variance. But I don't understand in which sense the bias and ...
0
votes
1answer
44 views

How to train neural network to say whether a given input belongs to the samples we have in dataset or not? [closed]

Suppose I have n samples with m features and a binary output (say 0 or 1). How do I create a neural network model which says whether a given input(to be predicted sample) is present in the samples(...
1
vote
0answers
52 views

Artificial Intelligence to Navigate websites [closed]

I have a requirement to create a bot that could Login to a website with credentials and then extract specific information from the website. Also the bot should be capable of doing the same on ...
0
votes
1answer
60 views

How to select activation functions for neural networks?

What is the process of selecting the best activation function? I already know which functions are better for which kind of problem, but I don't know why a particular function is better for a ...
0
votes
0answers
260 views

Epsilon Greedy Performing better than UCB

I am implementing the bandit problem using various algorithms. The issue that i am facing is that epsilon greedy is performing better than UCB for 5arms and horizon of 2000 for epsilon value of 0.95. ...
3
votes
1answer
80 views

What is an “undirected associative memory” in Hinton et al 2006?

In A fast learning algorithm for deep belief nets, the authors use the term "undirected associative memory". I am not sure what they are referring to, and unfortunately Google searches for this term ...
0
votes
2answers
37 views

Neural network trained with potentially incomplete data set

Please forgive my ignorance. I'm new to deep learning and this could prove to be a "stupid" question... I want to build a neural network that can predict some outcome "x". I have training data that ...
0
votes
0answers
20 views

Evaluation of a model in adaptive learning

I'm implementing an artificial intelligence for an adaptive learning software. We propose exercises to learn how to read. We have a first model that predicts the success or failure of students on the ...
0
votes
1answer
41 views

Big categorical data

I am trying to predict the price of used vehicles using three different models: Regression, ANN, and random forest. I am having 6 variables as an input for my model. One of my variables is the ...
2
votes
0answers
139 views

When to use batch normalization in a Fully Convolutional Network

I understand what batch normalization does but I'm not sure at what point it becomes overkill. With regards to a U-net, based on just the rough definition that batch normalization normalizes the ...
0
votes
0answers
35 views

closest meaning to a given word - NLP

I am trying to build a vocabulary tutorial. The tutorial will produce automated examinations for evaluating the user. In each of those exams, I want to show a word to the user and a form is there ...
0
votes
0answers
16 views

Help with starting a feedforward neural network for wind power forecasting

I am new to Machine Learning and Python and my task is to Predict wind power based on previous wind speed data. I have implemented this before using SVR. So I understand some basics. But now I'd like ...
0
votes
1answer
26 views

n * one_hot VS n_hot encoding for modeling input layer for a card game

How should I design my input layer for the following classification problem? Input: 5 cards in a card game; vocabulary is 52 cards Output: some classification using a neural network How should I ...
3
votes
1answer
23 views

Dynamic interactive learning

I am trying to solve a classification problem where I have a set of known X values. I know the classification objective i.e. the discrete set of values the Y can take. However, I don't have any ...
0
votes
1answer
163 views

machine-learning : Why training set and test set need to be independent and identically distributed?

My machine-learning book that I'm reading only says that they need to be but not why? My intuition says that if they are that leads to a better learning, if they were not it would be like we are ...
0
votes
0answers
31 views

Game problems that machine learning won't solve better than humans

Is there any type of game that could not be beaten by machine learning algorithm? Checkers and chess is not a problem for computers, go also.. But is there a class of problems in terms of let's say '...
0
votes
0answers
14 views

Using type-2-fuzzy sets and systems for automatic document summarization

I have already done a research project based on automatic document summarization using deep learning techniques. But fuzzy logic is pretty new to me, and I don't have any experience of building system ...
1
vote
0answers
29 views

Policy space optimization? Reinforcement learning

What are the reasons that you have an easier time choosing a Policy optimization algorithm when the state space is high dimensional, are there also any other reasons in the task's attributes for ...
1
vote
1answer
76 views

How can I interpret the following learning curves?

I am training a SVM with linear kernel over a training set of 3759 elements. The dimension of my problem is 2055, in other words, each example belonged to my training set is described by 2055 features....
0
votes
1answer
31 views

Create loss function thats takes probability in calculating

I am training object detection, the output is 5 numbers, probability that something is in the box , xmin, ymin, xmax , ymax What i am trying to achieve, is the "probability", to be scaled between 0 ...
3
votes
0answers
53 views

Find Feature Weighting in Deep Learning

If I train a deep neural network on standard tabular data (csv file etc. with labeled features) is there a good way to gauge how important each feature is in a particular new instance's prediction ...
1
vote
1answer
57 views

In MDPs with deterministic actions, should I use Q-learning or TD(0)?

Suppose in an Markov Decision Process (MDP), we have transition $(s, a, r, s', a', r', s'', ...)$, learning rate $\alpha$ and discount factor $\lambda$. The update formula of $TD(0)$: $V(s) \...
1
vote
2answers
68 views

Why this Way of iteration in RNNs? [closed]

One of the things that make them super-slow in training, is that they often use two nested loops to iterate on data through all time steps for every single iteration, calculating loss and cost at the ...
1
vote
1answer
209 views

Why can't experience replay be used for on policy methods if you re-sample new actions?

Suppose in your replay buffer you store only $(s_t, a_t, r_{t+1}, s_{t+1})$. Instead of finding the optimal Q function with this Bellman backup $Q(s_t, a_t) \leftarrow Q(s, a) + \alpha [R_{t+1} + ...
1
vote
1answer
90 views

Design of CNN whose input is a 2D game board

I am writing an artificial intelligence of a game. The algorithm needs a neural network to represent some information of the game board. The board is an 8*8 grid and each position in the grid has 4 ...
0
votes
4answers
2k views

Why is weight initialized as 1/sqrt(# of hidden nodes) in neural networks

I am currently reading Make your Own Neural Network by Tariq Rashid. He explains that instead of choosing weights randomly at a range of -1.0 to 1.0, initial weights should be in the range $ {1 \...