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.

Filter by
Sorted by
Tagged with
0
votes
0answers
10 views

Possible reasons that validation recall is fluctuating across different epochs but the precision is stable?

I know this is not a coding question, but didn't have any idea where can I ask for help on this. I'm training a deep learning model. After each epoch I measure the performance of the model on ...
0
votes
0answers
3 views

PLSR algorithm and terms like weight, loading, score

Can anyone suggest good resources to understand Partial Least Square Regression(PLSR) algorithm and the terms associated with it like weight, score, loading. I am doing a project on soil macro-...
0
votes
1answer
26 views

What is temporal leakage?

So I've been trying to work out exactly what temporal leakage is for a while now and I'm getting nowhere. I'm not necessarily looking to code or anything, I'm more so interested in what it actually is ...
0
votes
0answers
14 views

Dueling DQN - solving the identifiability problem

I want to implement a Dueling DQN. I know that the idea is to estimate V (scalar), the value of a given state, and A (vector) - the advantage, separately. A is the difference between V and the Q ...
2
votes
2answers
47 views

What could be an intutive understanding of a hyperplane?

This "Hyperplane" word gradually becoming more important to understand as I delve deeper into machine learning applications. To explain Hyperplane, the wiki article majorly speaks about one ...
0
votes
0answers
13 views

Re-training deep learning models multiple times to be able to compare their performance?

Let us say I have two deep learning models that differ in their hyperparameters and I want to compare their performance to each other (in terms of acc/ROC for example). However, a single value from ...
0
votes
0answers
11 views

Difference between CRF and Fully Connected CRF?

Can anyone explain me the difference between Conditional Random Fields and Fully Connected Conditional Random Fields for semantic segmentation? I only understand so far, that with CRF you try to use ...
0
votes
1answer
22 views

How do you combine an image and meta data as input in machine learning?

Well my question is really about text, but image is easier to understand. Let’s say I have some images and some meta data about the images and I want to train a model with the metadata and the image ...
0
votes
0answers
11 views

Choosing the sample size and number of samples for the bootstrap method to assess ANNs

Reading some online instructionals on the bootstrap method as a method to assess the performance of an Artificial Neural Network: https://machinelearningmastery.com/a-gentle-introduction-to-the-...
1
vote
0answers
11 views

How to compare the behaviour of different pretrained RL models?

I have a number of pretrained RL models (PPO2, ACER, ACKTR, ...) and I want to compare their behaviour in the environment. This includes their performance in respect of the episode-reward as well as ...
1
vote
1answer
54 views

Is there Automatic Selection of what Statistical Method Should be Used for Data Analysis?

I'm wondering if there exists a software/library that asks you questions about your dataset and then spits out the appropriate statistical method to use. I have searched online to find mostly that it'...
0
votes
0answers
4 views

Constraint optimization when target value is categorical

I am trying to build prescriptive model for feature parameter optimization. My dataset have both continuous and categorical features (x1, x2,x4,x4) and my target (y) is {0, 1} . I have build random ...
0
votes
1answer
16 views

How to train a model with time series to predict an object's weight?

Let's say I have an object, like a ball and I want to predict its weight. I have a bunch of different sensors that record a time series of the event. Each event is a different ball I would like to ...
1
vote
0answers
29 views

AI algorithm blind test - sample size

My company works in healthcare consulting, and are working with medical practitioners to explore how AI might be able to assist the remote, early diagnosis of central nervous system disorders such as ...
0
votes
1answer
27 views

Weighted importance sampling (WIS) and Importance sampling (IS)

I am currently reading papers about off-policy evaluation (or counterfactual evaluation) of reinforcement learning policies, including ones about the doubly robust estimator. As in this paper https://...
1
vote
0answers
17 views

when is it better to use One Vs All and when to use One Vs One in Statistics and Machine Learning?

I would love to further understand the uses and the application because while I understand the difference I still don't know which one is better for which type of problem.
0
votes
1answer
17 views

approach to classify text with natural language processing methods

I have a problem with regards to text classification/categorization. The task is bugging me for days already and as I am pretty new to AI and the field of natural language processing (NLP) I am just ...
7
votes
3answers
1k views

What happens in the sub-areas of AI? (ML, DL)

I have problems with understanding the sub-areas of AI and how it works. AI has the sub-area Machine Learning (ML), in which learning algorithms are used. Supervised/unsupervised learning takes place ...
1
vote
1answer
18 views

Is Matrix Factorization also going to work with one feature?

I need to fill missing values. I have found that there are many approaches such as the mean and the median of the feature as well as using Matrix Factorization. However, I am kind of wondering if I ...
0
votes
0answers
12 views

When is the manifold hypothesis violated?

If the high-dimensional dataset is of full rank, like an N samples by N features matrix, will the manifold hypothesis be violated? i.e. we cannot assume those data live on a lower-dimensional space?
0
votes
0answers
11 views

Xavier Initialization and Simoid activation function

One of the major assumptions made in Xavier initialisation is that the inputs to a neuron has a mean of zero. This is widely adapted in neural networks especially when the layers use tanh or sigmoid ...
0
votes
0answers
20 views

Should Training,test and validation data sets should be independent of each other in neural networks?

I have a dataset of n items and i draw three samples of equal size. These three samples are training set,test set and validation set. I want to use this data for training ,testing and to check ...
0
votes
0answers
24 views

Representation of state space, action space and reward system for RL porblem

I am trying to solve the problem of an agent dynamically discovering(start with no information about the environment) the environment and to explore as much of the environment as possible without ...
3
votes
1answer
27 views

Negative Log Likelihood cost

For multiclass classification, does the negative log likelihood loss function only take the loss for the classification group? i.e $$ C(\theta) \equiv \sum{}{}y_ilog(\hat{y}_i) $$ Doesn't $y_i$ just ...
0
votes
0answers
10 views

why do we forward pass the entire feature matrix and adjacency matrix in graph neural network training?

In GNN training phase, I saw some code (e.g. GCN, GAT) pass the entire feature and adjacency matrix forward and back propagate the train indexed features, and in test phase, they still pass the entire ...
5
votes
1answer
218 views

Empirical results of Machine Learning/Deep Learning in practice

The machine learning community often talks about over-fitting and other statistical issues. Are there any stats on how many models actually fail to reproduce good results after being deployed in ...
0
votes
0answers
20 views

How is the length of the output for encoder-decoder (seq2seq) models determined

I am trying to understand how encoder-decoder models works. The encoder receives a sequence and the length is known. However the output of the encoder is just a single word vector capturing the ...
0
votes
1answer
37 views

Correlation between features and the target

I am working on an AI project to predict the life time of an industrial tool. the data I have represents the consecutive Power values of the spindle during each use of the tool to produce a new piece. ...
0
votes
0answers
11 views

What does non parametric instance separation mean in this paper abstract?

I was reading the abstract of this paper - https://arxiv.org/pdf/1903.12355.pdf where they mention this paragraph in their abstract Unsupervised approaches to learning in neural networks are of ...
5
votes
1answer
71 views

Was Amazon's AI tool, more than human recruiters, biased against women?

A typical example how bias in data is being copied by AI is Amazon's recruiting tool that got abandoned in 2018. In the various reports it is implicitly (or sometimes explicitly) stated that the AI ...
0
votes
0answers
70 views

Implementing Actor-Critic with Experience Replay for Continuous Action Spaces

I have been trying to implement the ACER algorithm for continuous action spaces in reinforcement learning. The paper for the algorithm can be found here: Sample Efficient Actor-Critic with Experience ...
1
vote
0answers
143 views

Deductive reasoning and artificial intelligence

AI has proven to be extraordinary effective for solving certain types of intellectual problems that we thought before only our brains could solve. The number of applications is tremendous: engineering,...
0
votes
1answer
37 views

How can I draw decision boundaries for three different classes in Logistic Regression?

I understand the equation to draw a decision boundary in Logistic Regression with 2 independent variables and 2 different classes but when it comes to 3 different classes with 2 different independent ...
0
votes
0answers
43 views

Why Expectation and Maximization algorithm not used in Machine Learning while Gradient Descent algorithm used in Machine Learning?

I know that Newton Raphson, Expectation & Maximization, and Gradient Descent are all known to be optimization methods. Somehow, I wonder why Gradient Descent is chosen to be used in most of ...
0
votes
0answers
33 views

In feature selection, what the size of the data set is considered as too small? Is this an appropriate use of machine learning?

I am in a non-computer science field, and machine learning is being blatantly misused in my field. I recently got a journal paper to review, where the researchers used machine learning to develop a ...
0
votes
0answers
10 views

Can the same individual from a population be chosen in a genetic algorithm?

One population contains 10 individuals, from these 10, 8 are picked based on a selection mechanism. Let's say they are picked using the rank based selection mechanism (so the best ranked has a higher ...
2
votes
0answers
27 views

How does one solve the premise selection problem with imitation learning?

I was reading the following two new papers (HOList, Graph Representations) applying Machine Learning to Theorem Proving in Higher Order Logic. The main thing that I am unsure about is the following ...
0
votes
2answers
478 views

What is the difference between AI and “normal” programming?

Based on this link, DL is a subset of ML and ML of AI. My question is, is AI a subset of "normal" programming, or are they the same? The definition of AI from that link is: The broad discipline of ...
0
votes
0answers
24 views

Text Match for Description Fields

I have a long file with the list of item id's and their descriptions. These descriptions are like "The Brawn White Bolt Laser 10W" and "Laser for 10 Watt White Bolt" with different item id's. Though ...
1
vote
0answers
14 views

Mathematical proof of tradeoff between estimation error and computation cost in mini-batch gradient descent

" with more examples, the estimate would have a lower standard error, but the return is less than linear compared to the computational burden we incur." Came across this line while studying one ...
2
votes
1answer
112 views

What is the difference between Deep Neural Network(DNN) and Artificial Neural Network (ANN)? [duplicate]

This question basically focus on working of ANN and DNN. I really want to know, as both ANN and DNN may have multiple layer and also increase the number of hidden neuron. so, why DNN works better than ...
0
votes
1answer
36 views

What does State of Art Result means in context of ML/DL

Wondering what it means to have a state of art result.Is it a relative term or a standard? For Exmaple: If I have developed 2 models one with higher accuracy can i say i have achieved state of art ...
0
votes
0answers
25 views

Training a multi-layer perceptron (MLP) with a modified basis function

Consider a simple 3-layer MLP such as this. Each hidden layer implements y=xw+b where y is the output activation matrix of the ...
0
votes
1answer
17 views

Feeding multiple rows of data into ANN [closed]

I've built an ANN from scratch, that works with one row of data with any number of neurons and hidden layers. For the setup I am using 2 hidden layers, 5 neurons (just while building). The network ...
1
vote
0answers
17 views

Is it possible to reverse-engineer a variable which has a moderation effect in a multiple regression, given the other variables?

I am interested in simulating a hypothetical variable Z which moderates the relationship between X (predictor) and Y (outcome variable), while controlling for C (covariate). All variables are ...
1
vote
3answers
259 views

Predicting coefficients in a regular/polynomial equation using neural networks

Any insights onto how one can predict coefficients in lets say a regular or a polynomial equation using machine-learning/neural networks, i.e., $\alpha x_i + \beta x_i^2 + \gamma x_i^3 = y_i,$ ...
0
votes
0answers
32 views

How to understand 4 steps of Monte Carlo tree search?

From many blogs and this one https://web.archive.org/web/20160308070346/http://mcts.ai/about/index.html We know that the process of MCTS algorithm has 4 steps. Selection: Starting at root node R,...
1
vote
1answer
42 views

Policy Gradient Methods advantages over value-based methods

In the RL bible by Sutton and Barto it says on page 322 regarding the advantages of policy gradient methods: If the action space is discrete and not too large, then a natural and common kind of ...
0
votes
0answers
34 views

Is there a model or algorithm to improve digital drawings?

given a bad drawing, the algorithm should deform the edges of the bad drawing and fill them with color so that it looks more like a learned character. the entry would be a very poorly made drawing of ...
2
votes
0answers
56 views

Is studying Machine Learning really just learning a bunch of algos? [closed]

I've been watching some introduction Machine Learning videos just to see what it's all about and so far my takeaway has been that it just involves learning a bunch of algos (regression, k means, ...

1
2 3 4 5