Questions tagged [machine-learning]

Machine Learning is a subfield of computer science that draws on elements from algorithmic analysis, computational statistics, mathematics, optimization, etc. It is mainly concerned with the use of data to construct models that have high predictive/forecasting ability. Topics include modeling building, applications, theory, etc.

Filter by
Sorted by
Tagged with
1
vote
0answers
8 views

“Hand pose estimation in-the-wild” vs Normal hand pose estimation

So far I have seen several article that mention "Hand pose estimation in-the-wild" and just "Hand pose estimation." What is the difference between the two? Thank you for your help
2
votes
1answer
21 views

Given a regression based model with many feature variables; what tools would you utilize to figure out which feature variables add the most variance?

Given a hypothetical dataset {S} with 100 X feature variables and 10 predicted Y variables. X1 ... X100 Y1 .... Y10 1 .. 2 3 .. 4 4 .. 3 2 .. 1 Let's say I want to improve the accuracy of Y1. I am ...
0
votes
0answers
16 views

CNN Model Seems To Just Be Guessing

I am working with a binary classification problem, and regardless of what changes I make, the model seems to just be guessing between 0 (Negative) and 1 (Positive). The dataset is imbalanced at a ...
1
vote
1answer
27 views

On what principle did Google's DeepMind learn to walk?

I just saw this video on Youtube. On what principle did Google's DeepMind learn to walk? Was it Q-Learning or a Genetic Algorithm or Policy Gradient?
0
votes
0answers
13 views

What are some of the things that Policy Gradients Method can solve?

What are some of the things that Policy Gradients Method can solve that other Methods (like Q Learning and Genetic Algorithms) can't solve?
1
vote
0answers
8 views

Hierarchical multi-label classification problem using Neural Networks

What are the best practices/architectures to solve a Hierarchical multi-label classification problem using Neural Networks?
4
votes
1answer
180 views

How can precision be less than one in Leave-One-Subject-Out binary classification if each subject contains only one class

Say I'm trying to classify a medical condition. Theres only two classes: Sick and Healthy. I build a model and I can't split the data because I don't want data from the same patient being in training ...
0
votes
0answers
8 views

Using Wasserstein loss function for image-to-image-regression

The context I have a 3D array (representing a grayscale 3D image) and want to turn this into another 3D array of the same size. In this output array the value of each pixel is a number that measures ...
1
vote
1answer
32 views

Input layer is incompatible even when dimensions (apparently) match

I am making a sequential neural network for classification, with 3 dense layers, which will be trained on a simple synthetic dataset. The description of dataset is as follows: Data and class labels ...
2
votes
1answer
27 views

Dimensionality of the target for DQN agent training

From what I understand, a DQN agent has as many outputs as there are actions (for each state). If we consider a scalar state with 4 actions, that would mean that the DQN would have a 4 dimensional ...
3
votes
0answers
50 views

Unbiasedness of random forests

Suppose that I am trying to build a random forest by subsampling the data and choosing a single feature per tree randomly. For example, suppose there is some dataset, $D = \{(x_{1},y_{1}), ......(x_{N}...
0
votes
0answers
21 views

What Math is required to learn Policy Gradients (Part of Reinforcement Learning)?

That seems a lot of Math... So, in order for me to understand it... What topics of math should I learn? Or to summarize: What is the math prerequisite to learn Policy Gradients?
1
vote
1answer
56 views

How do I calculate the accuracy for graph mining in terms of (top 1%)?

I have 3600 samples in my dataset. I split the dataset into the train (2700) and test (900). My problem is related to ...
0
votes
0answers
11 views

Relative scoring of entities

I'm storing a set of house data in my Elasticsearch database with various attributes for each entity (such as price, number of bathrooms, sqft, etc..) I want to create a basic ranking engine that ...
0
votes
0answers
14 views

Why the prediction of this Random Forrest model is so poor in this machining dataset?

I am using Random Forrest to predict the MRR (Material removal rate). But the predictions have been quite off the mark. Even Linear Regression gave a much better result. I don't know where I'm going ...
0
votes
0answers
5 views

Binary classification or Single-class classification for data with Boolean label

In supervised learning. For simple prediction/classification problems like Will it rain tomorrow? or more serious one like disease diagnosis I often encountering ...
0
votes
0answers
10 views

How to retrain classification models

I have training data that is a set of transaction descriptions and the category of that transaction. I am trying to build a spend analyzer using this information to help me classify my transactions ...
0
votes
1answer
33 views

Drug Making Using Genetic Algorithms

I want to create a drug using N different chemicals for fighting a bacterial infection those N chemicals are contained inside the drug in different quantities my work environment is a simulated one ...
0
votes
1answer
11 views

reliability of human-level evaluation of the interpretability quality of a model

Christoph Molnar, in his book Interpretable Machine Learning, writes that Human level evaluation (simple task) is a simplified application level evaluation. The difference is that these experiments ...
0
votes
0answers
10 views

Handle unbalanced data by implementing Edited nearest neighbors, smote and Tomek links in r?

Imbalanced data is a big problem in classification problems. I have a binary classification problem with imbalanced data. I have researched and found that a possible method of dealing with this is ...
0
votes
0answers
14 views

I need help understanding RNN and toxic classification

How can an RNN be used for detecting toxic spans (spans of words containing toxic language) in a social media comment? Specifically, what should be the input to the RNN at each time step t? How many ...
0
votes
0answers
28 views

I don't understand the probability of PAC learning

I'm studying machine learning theory and I'm struggling with the concept of Probably Approximately Correct learning or PAC learning. The book I'm studying from makes an example where the hypothesis ...
0
votes
1answer
35 views

Would a neural network trained on extracted features have the same accuracy as a full network with frozen layers?

Let's say that I train two neural networks on the exact same dataset. The first network is a VGG19 model with frozen convolutional layers so only the top dense ...
0
votes
0answers
25 views

Which applications can not be handled by very Deep CNN models?

I wanted to know what challenges very deep models can face even if the accuracy is good. Would they be not suitable for any application given that my model is very very deep? I wanted to know if ...
1
vote
0answers
29 views

Dummy Variables of Weights in RNN Backpropagation Through Time

In the deep learning book RNN chapter (https://www.deeplearningbook.org/contents/rnn.html), it is mentioned that - To resolve this ambiguity, we introduce dummy variables $W^{(t)}$ that are defined to ...
0
votes
1answer
24 views

Prevent model from over-focusing on strong features

I have a classification model (DNN/Linear layers with some transformers and other things later). The input to the model are several different modalities of different lengths and different amounts of ...
0
votes
1answer
9 views

How to best visualize or capture time interval between lab measurements?

I have a table like as shown below ...
0
votes
0answers
6 views

Multi-object detection within single image

Given an image with multiple objects within it, I would like to train a CNN to output vector of labels corresponding to the presence/absence of objects within the image. I would like to know whether ...
0
votes
1answer
12 views

How to aggregate features to a group level as a feature in machine learning model?

I am building a model to predict some behavior at a household level. I could roll up income or number of cars etc so that I can take everyone into consideration. But how can I roll up something like ...
0
votes
0answers
6 views

Speed up convergence of GLM fitting in custom GLM code (using GD)

So, I've got a little bit of an unusual situation. I do ML on encrypted data, and the ecosystem there is relatively young, so I have to build my own infrastructure, and there are certain restrictions ...
0
votes
0answers
16 views

Why do relative positional embeddings instead of absolute positional encoding improve the Transformer?

I've been researching the Music Transformer, the paper for which introduced an efficient algorithm to compute Relative Positional Embeddings in a Transformer. I know that Relative Positional ...
0
votes
0answers
10 views

Selecting which categorical variable optimizes a target variable?

this is my first question here! I am working on developing a machine learning model that can select which NFL play type (pass, run, punt, or kick) will optimize win probability added based on a ...
-1
votes
0answers
14 views

What is the difference between statistical learning and deep learning?

statistical learning and deep learning are the sub field of machine learning. But what is the core difference between statistical learning and deep learning and where do they both intersect ? If they ...
1
vote
0answers
26 views

Which samples to inspect?

I have a dataset with noisy labels on which I train a binary classifier. Inspecting the loss I see some samples were misclassified with high confidence and others were classified with indecision ...
0
votes
2answers
12 views

How to interpret confusion matric for Extreme learning machine ? why it has decimal points in it?

ELM for phishing URL detection. Dataset is divided into 70/30. 30% is for testing. Dataset size: 11,000 Entries.
1
vote
1answer
35 views

with ML/DL model Is possible predict numbers of items required?

I have a dataset is regarding ambulance call data. Data sample: ...
-2
votes
2answers
43 views

Recent comprehensive text-books on machine learning? [closed]

I would like to ask what are the current recommendations for professional and comprehensive, current books on machine learning. The two quite classic comprehensive text-books are: Machine Learning: A ...
1
vote
0answers
10 views

When are subword ngrams trained in fasttext? (Enriching Word Vectors with Subword Information)

when is the training for subword ngrams done? is it done simultaneously as when the word representation are trained? or is it done at the end, after word representations are created? fasttext ...
0
votes
0answers
18 views

What is the Intuition behind weight vector W which is normal to the plane? Is the weight vector W same as the W which is normal to the plane π?

In an interview, I was asked the intuition behind the weight vector. I told the weight vector is a vector which we try to minimize to a local minima with the help of regulariser so we don't overfit. ...
1
vote
0answers
10 views

What is a good reward function when objective is to minimize the average along with the variance?

I am trying to formulate a problem where we are trying to minimize the average resource allocated to different users. Due to some inherent properties of the environment, some users can be easily ...
1
vote
1answer
21 views

Need help understanding Hard SVM quadratic program equation

This is from the textbook "Understanding Machine Learning" by Shalev-Schwarz p. 169. Can anyone help me understand why the solutions to this optimization problem need to be divided by the ...
0
votes
0answers
11 views

How could I find the number of instances per fold to plot a histogram, in my dataset?

I am currently trying to learn machine learning with Python, and I have been given an .npz file containing the data. I have explored the dataset, and after exploration noted that it has 5 datatypes: <...
0
votes
0answers
11 views

why positive loss is going up in binary classification problem?

I am training a model for object detection with Lidar using binary cross entropy. But my positive_class_loss(foreground) is going up, whereas negative_class_loss is going down(background). But ...
1
vote
1answer
25 views

Mathematical bias and weight vs machine learning bias and weight

I am a little confused about the term Bias and Weight with respect to machine learning. Say we want to predict the heights of people whose weights are given. So plot weights to x-axis and height to ...
0
votes
1answer
33 views

Does REPEATED K-fold cross validation make sense with Random Forest? [closed]

When using random forest, would using normal cross-validation and just taking the average results from multiple models with different random states give me the same results as using Repeated K-fold ...
0
votes
0answers
20 views

Predicting % of demand going to each product

I work within an industry with products that expire, therefore we would like to be able to choose which specific marketing keywords we should switch on to drive demand to the products that are over-...
1
vote
1answer
25 views

What is the intuition of using clustering for performing feature engineering in machine learning tasks?

I am trying to implement the research paper Combining Boosted Trees with Metafeature Engineering for Predictive Maintenance. The paper has a section called meta feature engineering where they have ...
0
votes
0answers
6 views

How to do a t-test on rare events distributions?

I have a dataset, which contains several features (feature_1, feature_2, ..., feature_n) and ...
1
vote
1answer
16 views

Which algorithm would be suitable for clustering a billion datapoints?

I am running a K-means algorithm (using the sklearn implementation) on an aggregated dataset of ~350k datapoints on a 6 dimension hyper-plane (using 6 features). I ...
1
vote
0answers
11 views

The best practice for implementing a dynamic classification model that can have classes added to it frequently?

Let's say you are tasked with training a malware classification model, but the problem is that new malware families could be added to this model every week. for example a new malware family shows up, ...