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 [tsne]

T-distributed stochastic neighbor embedding (t-SNE) is a nonlinear dimensionality reduction algorithm introduced by van der Maaten and Hinton in 2008.

1
vote
2answers
316 views

Intuitive explanation of how UMAP works, compared to t-SNE

I have a PhD in molecular biology. My studies recently started to involve high dimensional data analysis. I got the idea of how t-SNE works (thanks to a StatQuest video on YouTube) but can't seem to ...
9
votes
1answer
121 views

t-SNE with mixed continuous and binary variables

I am currently investigating the visualisation of high-dimensional data using t-SNE. I have some data with mixed binary and continuous variables and the data appears to cluster the binary data much ...
2
votes
2answers
227 views

Why does larger perplexity tend to produce clearer clusters in t-SNE?

Why does larger perplexity tend to produce clearer clusters in t-SNE? By reading the original paper, I learned that the perplexity in t-SNE is $2$ to the power of Shannon entropy of the conditional ...
4
votes
1answer
160 views

How can t-SNE or UMAP embed new (test) data, given that they are nonparametric?

I have started using the UMAP method for dimension reduction which is a similar method to t-SNE, Diffusion Maps, Laplacian Eigenmaps, etc. The named dimension reduction methods have in common that ...
2
votes
1answer
53 views

Perplexity formula in the t-SNE paper vs. in the implementation

The perplexity formula in the official paper of t-SNE IS NOT the same as in its implementation. In the implementation (MATLAB): ...
1
vote
0answers
154 views

How do I intepret these t-SNE results?

I am rather new to the TSNE method, and am learning about the various pitfalls associated with interpreting it correctly. I have performed TSNE on a very-high dimensional dataset (>20,000 dimensions) ...
1
vote
1answer
87 views

How is Student's t-distribution related with this similarity/probability equation between data points?

In the t-SNE paper "Visualizing Data using t-SNE" and a Deep Embedded Clustering (DEC) approach "Unsupervised Deep Embedding for Clustering Analysis", they both use the Student t-distribution to ...
1
vote
1answer
87 views

Why does modelling t-SNE 2D coordinates with a neural network fail to produce a perfect fit?

In short, I want to train a neural network to project my high dimensional objects to a 2D plane coordinates which was created using t-SNE (Rtsne package for R). I ...
6
votes
1answer
119 views

Why can't t-SNE capture a simple parabola structure?

As a toy example, I used t-SNE on a simple parabola to have a representation of it in one dimension. ...
2
votes
0answers
36 views

Does t-SNE in 3 dimensions offer any additional information to 2 dimensional t-SNE?

t-SNE is a data reduction technique that primarily gives information about local similarities. Clustering in a t-SNE graph usually (but not always) indicates some hidden structure in the higher-...
0
votes
1answer
35 views

Is there a way to reduce high-dimensional feature space to an array of 2d tSNEs ordered along a chosen dimension?

Let's say we have 4096-d vectors (via a CNN fully-connected layer) and often we use tSNE to visualize the space, sometimes in combo with Jonker-Volgenant to assign it to a grid. When applied to image ...
1
vote
1answer
191 views

Can t-SNE help feature selection?

I'm training a fully connected feed forward neural network for regression. Given one training example $(x_i, y_i)$, I need to convert the raw representation $x_i$ into an invariant representation $...
0
votes
0answers
22 views

What distinguishes the different dimensions of t-Distributed Stochastic Neighbor Embedding (tSNE)

I understand conceptually how the tSNE algorithm works in one dimension. But I am confused how this works for 2 or more dimensions (let us call the number of dimensions of the lower dimensional space ...
0
votes
0answers
153 views

Using UMAP or other non-linear dimension reduction techniques on response variables prior to learning?

Background Suppose you have a training set where the response measurements are some $N$-dimensional vectors of related measurements - in my specific case, they happen to be cell viability scores for ...
1
vote
0answers
106 views

Log-transformation of Compositional Data

I am dealing with compositional data, in a high dimension. Each sample I have behaves like: $$ {S}^D=\left\{\mathbf{x}=[x_1,x_2,\dots,x_D]\in\mathbb{R}^D \,\left|\, x_i>0,i=1,2,\dots,D; \sum_{i=1}...
1
vote
1answer
65 views

Where is t-distribution used in t-SNE?

I am trying to learn the dimensionality reduction using t-SNE technique. After some videos and explanation I understood the idea behind it. But I am not getting where the t-distribution is used behind ...
1
vote
1answer
45 views

Interpreting t-SNE dimensions in terms of the original features [duplicate]

I have a dataset of 223 multilabels. I'm sure some of those labels are correlated so I want to merge them. For that I'm using sklearn tSNE to reduce the dimensions. How can I "describe" the new ...
2
votes
0answers
142 views

Intuition behind perplexity parameter in t-SNE

While reading Laurens van der Maaten's paper about t-SNE we can encounter the following statement about perplexity: The perplexity can be interpreted as a smooth measure of the effective number of ...
1
vote
2answers
49 views

Dimensionality reduction: include labels?

Lets say I have something like the iris dataset, the columns are petal length, petal width, ...
0
votes
1answer
83 views

Dimensionality reduction with least distance distortion

Question: Could I find a dimensionality reduction algorithm without or with minimal distance (cosine) distortion? Background: I would like to visualize in 2D a sample of news texts for which I also ...
3
votes
0answers
176 views

Differences between t-SNE and SOM

I have some high dimensional data and I want to reduce it to 2 dimensions for visualization. The goal is to color the points in this 2D space to see whether there is any clustering due to different ...
3
votes
1answer
120 views

Can we apply KL divergence to the probability distributions on different domains?

When I was reading the original paper of t-SNE, I had an question whether or not we can apply KL divergence to the discrete probability distributions on different domains. In the paper, they measure ...
0
votes
0answers
96 views

t-sne V CNN extracted features

If one were to visualise images (say MNIST) with t-sne we get great separation in the t-sne lower dimensional space. However what would happen if one where to run a CNN on MNIST, then remove the ...
2
votes
1answer
194 views

t-SNE on principal component scores: standardization needed?

I have a huge dataset (1.5 million obs and 70 features). I want to visualize the data in 2D, to look for naturally occurring clusters. Analogous to Van der Maaten's approach 1, I first reduce the ...
1
vote
1answer
147 views

Can I reconstruct the original data from a dimensionality reduced data using t-SNE?

I have implemented dimensionality reduction using PCA, AE, and t-SNE! I am trying to compare the output of the three methods by finding the reconstruction error. I have managed to reconstruct the data ...
3
votes
3answers
2k views

Are dimensionality reduction techniques useful in deep learning

I have been working on Machine learning and noticed that most of the time, dimensionality reduction techniques like PCA and t-SNE are used in machine learning. But, i rarely noticed anyone doing it ...
8
votes
1answer
162 views

How is the family of distributions with PDF proportional to $(1+ax^2)^{-1/a}$ called?

Consider a family of distributions with PDF (up to a proportionality constant) given by $$p(x)\sim \frac{1}{(1+\alpha x^2)^{1/\alpha}}.$$ How is it called? If it does not have a name, how would you ...
2
votes
1answer
45 views

Visually distinct clouds, but stuck together

I have 96 features and the labels are represented by 1 and -1 for inputting to a deep learning model. 1- PCA Here the 3 axis represent the 3 first principal components. The blue cloud represents the ...
3
votes
2answers
320 views

Why Kullback-Leibler in Stochastic Neighbor Embedding

Stochastic Neighbor Embedding (and t-SNE) relies on Kullback-Leibler divergence between the point distributions in the original and the low-dimensional space. Why? Why not any other dissimilarity ...
1
vote
0answers
26 views

Developing novel association measure with t-SNE

I am an MSc. student in computing and I am currently writing my thesis on radiogenomics (imaging genomics) for brain cancer research. I NEED YOUR HELP. My supervisor and I are thinking about ...
3
votes
1answer
75 views

The usage of “variance” (vs. “standard deviation”) in the 2008 t-SNE paper

In the t-SNE paper from 2008 the term "variance" is mentioned on several occasions. For example on page 3: For nearby datapoints, $p_{j|i}$ is relatively high, whereas for widely separated ...
1
vote
2answers
720 views

How to do dimension reduction with sparse data

I have 200 vectors representing the percentage marks for 200 different students in the different classes they took. The vectors are 22 dimensional (as there were 22 different classes in total) even ...
0
votes
1answer
857 views

Does it make sense to run DBSCAN on the output from t-SNE? [duplicate]

Performed time series clustering where I used DTW to generate a distance matrix. The distance matrix was then given as an input to t-SNE where the two-dimensional results from t-SNE were used for ...
12
votes
1answer
786 views

What is the meaning of the axes in t-SNE?

I'm currently trying to wrap my head around the t-SNE math. Unfortunately, there is still one question I can't answer satisfactorily: What is the actual meaning of the axes in a t-SNE graph? If I were ...
2
votes
1answer
51 views

What is meant by: neighbor identities are preserved in t-SNE?

From the article on Stochastic Neighbor Embedding (by Geoffrey Hinton and Sam Roweis): Stochastic Neighbor Embedding (SNE) tries to place the objects in a low-dimensional space so as to ...
0
votes
0answers
45 views

Interpretability of the dimensions in t-SNE [duplicate]

I'm relatively new to t-SNE and it seems that unlike in other dimensionality-reduction techniques, the dimensions in t-SNE are hard to interpret. The contribution of the variables can easily be ...
2
votes
1answer
503 views

t-SNE, PCA and another technique - which one? [closed]

I am comparing dimension reduction techniques and I am utilizing them for data visualizations onto a plane – projections in 2D space. The input into a projection/dimension reduction techniques is a ...
0
votes
1answer
131 views

t-SNE vs mm t-SNE [closed]

I have a question concerning t-SNE algorithm implementations in R. I have a dataset whose dimensionality is about 300, and I want to visualize it in 3 dimensions. I tried the ...
1
vote
0answers
21 views

Denominator in pairwise similarities in symmetric SNE

I am trying to understand how symmetric SNE (stochastic neighbour embedding) works while reading the paper of Van der Maaten about t-SNE. However, I am a bit confused about the denominator in the ...
29
votes
3answers
8k views

Why is t-SNE not used as a dimensionality reduction technique for clustering or classification?

In a recent assignment, we were told to use PCA on the MNIST digits to reduce the dimensions from 64 (8 x 8 images) to 2. We then had to cluster the digits using a Gaussian Mixture Model. PCA using ...
3
votes
1answer
67 views

What is meant by “scale” in the 2008 t-SNE paper?

I am trying to understand the use of the term “scale” in the 2008 van der Maaten and Hinton t-sne paper. I’m not sure I exactly understand what they mean by their use of the term “scale”, for ...
4
votes
1answer
351 views

Why does t-SNE not separate linearly separable classes?

I have small convolution neural net for classification EEG patterns (2 class problem) with several conv layers and one dense layer. This net performs pretty well. For education purposes I used ...
2
votes
1answer
374 views

Why do I get weird results when using high perpexity in t-SNE?

I played around with the t-SNE implementation in scikit-learn and found that increasing perplexity seemed to always result in a torus/circle. I couldn't find any ...
3
votes
1answer
1k views

How to interpret t-SNE plot?

I want to know how to interpret t-distributed stochastic neighbor embeding (t-SNE) plots. In particular: 1) What information do they convey, besides showing clusters? 2) In PCA we can see loadings and ...
1
vote
0answers
35 views

Classifier training long time due to the size data

I have a problem to train my classifier. I have 10 different kinds of music genres, each genre with 100 songs, after making an Mfccs I have a numpy array of (1293, 20) If all together with np.vstack I ...
2
votes
1answer
68 views

when not to decide a problem cannot be solved with ML

I am working with a typical ML problem, trying to separate 2 classes in a supervised manner. I wanted to ask at which point you decide the data is not descriptive enough to solve the problem, and if ...
0
votes
1answer
559 views

DBSCAN considers all data points noise for reduced time series data [closed]

I had a data matrix 609 rows × 264 columns, time-series data. Data was reduced using t-SNE algorithm to 3 dimensions. When being clustered I get zero clusters, where all data points are considered ...
6
votes
4answers
667 views

t-SNE dimensions as additional predictor variables

This question could also (maybe) relate to PCA. I built a supervised RandomForest on a dataset that I'm currently working on - the actual V Prediction $R^2$ was holding around 80% across many CV ...
0
votes
1answer
157 views

Using t-SNE to visualize the evolution of a vector field over time

I have a data set of 100 high-dimension matrices, each describing the state of a vector field at a different time. My goal is to use t-SNE to visualize how this vector field evolves over time. However,...
1
vote
1answer
74 views

Bandwidth of the gaussian kernels in t-sne

I'm trying to understand t-SNE better and I was hoping someone could elaborate on how the $\sigma _i$'s are chosen. I was also wondering why they aren't just calculated in the normal way standard ...