Questions tagged [machine-learning]

Machine learning algorithms build a model of the training data. The term "machine learning" is vaguely defined; it includes what is also called statistical learning, reinforcement learning, unsupervised learning, etc. ALWAYS ADD A MORE SPECIFIC TAG.

5,434 questions with no upvoted or accepted answers
Filter by
Sorted by
Tagged with
5
votes
1answer
2k views

From Markov Decision Process (MDP) to Semi-MDP: What is it in a nutshell?

Markov Decision Process (MDP) is a mathematical formulation of decision making. An agent is the decision maker. In the reinforcement learning framework, he is the learner or the decision maker. We ...
5
votes
0answers
10k views

True positive, false negative, true negative, false positive definitions for multiclass-multilabel classification?

I'm trying to apply some evaluation metrics to several clustering methods. I thought that I knew them basing on the multiclass confusion matrix, considering the rows as the actual classes and the ...
5
votes
0answers
2k views

Which standard deviation of the cross-validation score?

When doing cross-validation for model selection, I found there are many ways to quote the "standard deviation" for the cross-validation scores (here "score" means an evaluation metric e.g. accuracy, ...
5
votes
0answers
1k views

Recurrent neural network for object tracking & position filtering?

Would a recurrent neural network be appropriate for object tracking tasks? Mainly I will have 3D feature vectors $(x, y, t)$ where $x$ and $y$ are the positions of an object in the image and $t$ is ...
5
votes
0answers
352 views

Interpreting hidden layer representations in ANNs

I'm using the fann library for writing an Artificial Neural Network in C++. I trained my network for the task of recognizing faces inside a set of 128x128 .png images, using three different algorithms:...
5
votes
0answers
407 views

Ideal statistical or machine learning technique to model highly cross-correlated data

I'm trying to build a model that can predict streamflow for an alpine (snowmelt-fed) watershed using snow albedo (roughly, the energy reflectance of the snow) data. Albedo controls the melt of the ...
5
votes
0answers
365 views

What enforces features diversity in RBM?

I'm working on an implementation of a Restricted Boltzman Machine (RBM). I made some tests on the MNIST dataset trying to learn a representation of the digit 2. My inputs are binary images. My aim is ...
5
votes
0answers
510 views

Basic idea of zero inflated two part models(hurdel) and application to big data (machine learning)

I'm currently working on the data which has 90% 0s in response variable. Based on my research, it seems zero inflated models could be a solution to this. However, while I was reading related documents,...
5
votes
0answers
392 views

Rademacher bounds for unbounded loss functions

All common treatment of PAC bounds based on Rademacher complexity assume a bounded loss function (for a self-contained treatemnt, see this handout by Schapire. However, I could not find any result for ...
5
votes
1answer
1k views

Intuition behind perceptron algorithm with offset

I was looking for an intuition for the perceptron algorithm with offset rule, why the update rule is as follows: cycle through all points until convergence: $\textbf{if }\, y^{(t)} \neq \theta^{T}x^{(...
5
votes
0answers
151 views

Compressed sensing: Optimization in $L_1$ norm and total variation with fourier coefficients

I'm reading the article Robust Uncertainty Principles: Exact Signal Reconstruction from Highly Incomplete Frequency Information (Candes, Romberg and Tao, 2004). In this article they are talking ...
5
votes
0answers
430 views

Any implementations of fully recurrent neural networks applied to reinforcement learning?

I've seen a single paper on the topic of adapting fully recurrent networks to a reinforcement learning setting, but according to google scholar its had no citations and no code has been released ...
5
votes
0answers
193 views

How to learn similarity of typed/attributed graphs?

I have a question for graph machine learning gurus :). For this project I'm working on, I need to be able to learn similarity between typed graphs. By typed I mean that every vertex and every edge of ...
5
votes
0answers
125 views

Graphical nominal model

Suppose I have a set of $k$ matrices. $$ \mathbb D = A_1,A_2,...,A_k $$ Each column of $A$ is categorical vector. $$ A = v_1,v_2,...,v_n $$ I want to find the mapping $$ f: A \...
5
votes
0answers
181 views

Asynchronous data stream matching

Suppose you have a classifier $C^n$ which continuously outputs a stream of classification labels $K^n_i$ and corresponding timestamps $T^n_i$. Also, we know the prior probability $P(K^n) \forall n$. ...
5
votes
0answers
2k views

Why would concatenating feature vectors lead to better estimates?

I wish to estimate the state of a system from two separate and disparate observations. A simple approach that I have seen in some literature is to combine the feature vectors (observations) by simply ...
5
votes
1answer
647 views

Unlearning Neural Network? Prevent learning from a specific feature

Is it possible to train a NN to avoid the features that a different neural network finds? For example, let's train a simple 1 layer CNN with 1x1 kernels on a supervised binary classification problem. ...
4
votes
2answers
84 views

Reconstruction Error: Principal component analysis vs Probabilistic prinicpal component analysis

I am working through the book "Machine Learning: A Probabilistic Perspective". After introducing PCA and Probabilistic PCA, the following graphic is shown (the upper two graphics ...
4
votes
0answers
57 views

Why is BART so accurate in causal inference?

The famous paper Dorie,2017 shows that BART performs dramatically well in causal inference. In my replication, MSE in BART can be 40% lower than MSE in other machine learning methods. But all machine ...
4
votes
0answers
194 views

Understanding Object2Vec

AWS released an interesting feature as part of the SageMaker service called Object2Vec that lets you make an embedding for search out of pretty much anything: documents, users, products, ...
4
votes
0answers
48 views

Why is Rademacher complexity defined the way it is?

For reference, this is the definition of empirical Rademacher complexity from Foundations of Machine Learning (page 30): Let $\mathcal{G}$ be a family of functions mapping from $\mathcal{Z}$ to $[...
4
votes
0answers
71 views

How to explain random forest ML algorithm doesn't learn at all, while logistic regression learns very well?

My prediction task is as follows: Use name to predict people's ethnicity (into 4 categories: "English", "French", "Chinese", and "All others") as a multiclass classification problem. The name ...
4
votes
0answers
33 views

How is EMSE derived for causal trees in Athey and Imbens (PNAS 2016)?

Athey and Imbens build a non-parametric matching procedure to identify and estimate causal effects. To this end, they minimize the expected mean squared error (EMSE) of their procedure, but I don't ...
4
votes
1answer
73 views

Workaround for word embeddings that do not “see” antonyms

Most word embeddings do not “see” antonyms. For instance, among many words they will place vectors for “dependent” and “independent” (as an example) quite close, — actually as close as with synonyms ...
4
votes
1answer
433 views

How is TD(1) of TD(lambda) equivalent to Monte Carlo?

In Sutton and Barto's book about RL they say that the TD($\lambda$) algorithm is equivalent to Monte Carlo when $\lambda = 1$. I don't see how that is the case. They define the lambda return as: $$...
4
votes
0answers
166 views

The extrapolation problem: model selection, performance metrics, and improvement

Machine learning models are fit to a response variable within a given range. This leads to weak and sometimes disastrous performance when it comes to instances with an actual response variable outside ...
4
votes
0answers
104 views

How to model probabilistic inputs with continuous output using regression

I have trained a multi-output classifier that takes an image as input and returns softmax logits as output. To be specific, the multi-output classifier takes an image and says the probability that ...
4
votes
1answer
76 views

Why do we use the average-pooling layers instead of max-pooling layers in the DenseNet?

In the paper of DenseNet. The author adopted average-pooling in the transition layers. So what is the motivation of such choice? Why not using max-pooling layers?
4
votes
0answers
63 views

How to reconstruct an image from a training set?

Description: I have taken a series of images/photos of a panorama from different positions (x,y) in space pretty close to each other (max 100m difference). Here there is a top view representation to ...
4
votes
0answers
83 views

Why is this statistic F-distributed?

A book I'm reading claims that the statistic: $\frac{(RSS_0 - RSS_1) / (p_1 - p_0)}{RSS_1 / (N - p_1 - 1)}$ has an F distribution. Why is this? I know that an F distribution is something like $\frac{\...
4
votes
0answers
56 views

How can statistics be used to avoid “Lending False Credibility To Decisions We've Already Made”

In light of this article Data Science Has Become About Lending False Credibility To Decisions We've Already Made published in Forbes, I would appreciate input from the statistical and data science ...
4
votes
0answers
325 views

Analyse sensitivity of hyper-parameters of Machine Learning Models

I want to analyse how sensitive my non neural net machine learning models are to the choice of the different parameters. I am currently using grid search to tune the models. Is there any method that I ...
4
votes
0answers
226 views

PCA, SMOTE and cross validation - how to combine them together?

I was reading a lot recently about PCA and cross validation and it seems that the majority call it malpractice to do PCA before cross validation. I would also like to perform SMOTE, but there is a ...
4
votes
1answer
322 views

How to predict routes using clustering data

I've been working on a ship route prediction algorithm such that given the past and current trajectory of a ship I am able to estimate the future one. The trajectories are represented as a sequence of ...
4
votes
0answers
668 views

Why do we need the gamma parameter in the polynomial kernel of SVMs?

The polynomial kernel is sometimes defined as just: $$ K(x,y):=(\left<x,y\right>+c)^d $$ with two parameters: the degree $d$ and constant coefficient $c$. But others (e.g., libsvm, and sklearn ...
4
votes
0answers
306 views

Why researchers use conv1d for embeddings instead of dense layers?

In some papers (like Reinforcement learning for Vehicle Routing Problem), researchers use conv1d to embed the problem input into a hyperspace; for example, in solving TSP, they use conv1d on the (x,y) ...
4
votes
0answers
241 views

May too much batch normalization hurt learning?

I was experimenting with some CNN models and reading research material when I realized that it could happen that using only a single batch normalization layer at the early stages of the network could ...
4
votes
0answers
62 views

Inferring a Markov chain from its invariant measure

Given a probability measure $p$ on $\{1,\dots,n\}$ assumed to be the invariant measure of some irreducible ergodic Markov chain with unknown transition matrix $P$, i.e., $p = pP$, what (if any) ...
4
votes
1answer
610 views

What does it mean if the ROC AUC is high and the Average Precision is low?

I have a model that produces a high ROC AUC (0.90), but at the same time a low average precision (0.30). From what I've found, I think it might have to do something with imbalanced data (which the ...
4
votes
0answers
229 views

Correcting Sample Selection Bias given actual Distribution

I have two datasets, both from the same population: The samples from the first survey are quite representative of the underlying truth. However, the second survey comes with a change in distribution ...
4
votes
0answers
74 views

soft SVM - degenerate case

According to "A Note on Support Vector Machine Degeneracy", Theorem 4, if the dual problem for soft-SVM has a solution with $\alpha_i \in \{0,C\}, \forall i$, then $w=0$ for the primal problem. In "...
4
votes
1answer
61 views

Gradual clustering in deterministic manner

We have 128-dimensional vectors representing people's identities where the euclidean metric defines the similarity between them. Ours solution requires them to be clustered and then annotated (assign ...
4
votes
1answer
37 views

How do these matrices form an order-$4$-tensor?

I'm reading this paper on a convolutional neural network for modelling sentences, and I'm having some trouble understanding section $3.5$. Please consider the following text: We denote a feature map ...
4
votes
0answers
194 views

Why do we use gradients instead of residuals in Gradient Boosting?

I have found mentions of two advantages in using gradients instead of actual residuals: 1) Using gradients will allow us to plug in any loss function (not just mse) without having to change our base ...
4
votes
0answers
497 views

Position Bias Normalization in CTR Prediction

I am working on Click Through Rate(CTR) prediction model on a toy dataset. The label I am using is #Click/#NumShown. But there is position bias in results shown. ...
4
votes
0answers
74 views

How does one design a data set from polynomial target function such that logistic regression separates the data perfectly?

I want to design a target function for a classification task of the form: $$ f_{target}(x) = \mathbb{1}_{>0}[\sum^{D^*}_{i=0} w^*_i x^i] = \mathbb{1}_{>0}[ \langle w^*, \Phi(x)\rangle]$$ and ...
4
votes
0answers
619 views

Effectiveness of Standardization and Normalization in Machine Learning

I am just studying the basics of machine learning and had a question about the standardisation and normalisation of the features and its effectiveness. I have read this CrossValidated question and ...
4
votes
0answers
49 views

using sparse models to accelerate training

I was going through a tutorial that highlights the importance of using sparse models to get "lightning fast models" i.e. to accelerate the training process. What are the issues in using a sparse ...
4
votes
0answers
632 views

Supervised Learning: Cosine Similarity as Loss function?

For supervised learning, the loss function should be differentiable so that back-propagation can be performed. I am wondering if it is possible to use loss function that computes the cosine similarity?...
4
votes
0answers
957 views

SVM Primal Optimization and Dual Optimization. Under what cases one better than another?

Primal $$argmin_{w} \frac{1}{2}||w||^2 + \frac{1}{N}\sum max(1-t^{(i)}(w^Tx+b),0)$$. The dual form $$max_{\alpha} \sum\alpha_i - \frac{1}{2}\sum t^{(i)} t^{(j)}\alpha_i\alpha_j K(x^{(i)},x^{(j)})$$...