There is a very simple argument that shows that MDS and PCA achieve the same results.

This argument has 2 important components. The first of these shows that an eigendecomposition of a gram matrix can be used for dimension-reduction.

PCA leverages the singular value decomposition (SVD). Given a matrix $ X $, the SVD is $ X = USV^{T} $.

We drop columns from X by using $ US_{t} $ where we drop some rows and columns from S.

This is also conviently obtained using an eigendecomposition of the covariance matrix.

Working with the gram matrix, we have $ XX^{T} $ and when expressed in terms of $ U $, $ S $ and $ V $, we have $ XX^{T} $ = $ (USV^{T})(VSU^{T}) $.

Simple algebra tells us that this is equal to $ US^{2}U^{T} $. The spectral theorem tells us that this the eigendecomposition of the gram matrix will return this decomposition. $ U $ and $ S $ can be retrieved and a dataset with fewer dimensions can be obtained.

The second part of the argument involves proving that a matrix of distances is indeed a gram matrix. This argument was discussed in a previous post.

Last week, I dropped out of the PhD program at LTI, SCS, Carnegie Mellon University with a masters degree. The specifics of this decision will remain private. If you are going through issues in your PhD program, I will be happy to talk to you and give you advice (should you desire any). You can email me at: shriphanip@gmail.com. You can also call me at 425–623–2604.

Eigendecompositions and Singular Value Decompositions appear in a variety of settings in machine learning and data mining. The eigendecomposition looks like so:

$$ \mathbf{A}=\mathbf{Q}\mathbf{\Lambda}\mathbf{Q}^{-1} $$

$ \mathbf{Q} $ contains the eigenvectors of $ \mathbf{A} $ and $ \mathbf{\Lambda} $ is a diagonal matrix containing the eigenvalues.

The singular value decomposition looks like:

$$ \mathbf{A} = \mathbf{U} \boldsymbol{\Sigma} \mathbf{V}^* $$

$ \mathbf{U} $ contains the eigenvectors of the covariance matrix $ \mathbf{A}\mathbf{A^T} $. $ \mathbf{V} $ contains the eigenvectors of the gram matrix $ \mathbf{A^T}\mathbf{A} $.

The **truncated** variants of these decompositions allow us to compute only a few eigenvalues(vectors) or singular values (vectors).

This is important since (i) a lot of times, the smaller eigenvalues are discarded, and (ii) you don’t want to compute the entire decomposition and retain only a few of the rows and columns of the computed matrices each time.

For core.matrix, I implemented these truncated decompositions in Kublai. Details below.

*This is part of a series on a family of dimension-reduction algorithms called non-linear dimension reduction. The goal here is to reduce the dimensions of a dataset (i.e. discard some columns in your data)*

In previous posts, I discussed the MDS algorithm and presented some key ideas. In this post, I will describe how those ideas are leveraged in the Isomap algorithm. A clojure implementation based on core.matrix is also included.

In a previous post, I described the MDS (multidimensional scaling) algorithm. This algorithm operates on a proximity matrix which is a matrix of distances between the points in a dataset. From this matrix, a configuration of points is retrieved in a lower dimension.

The MDS strategy is:

- We have a matrix $ D $ for distances between points in the data. This matrix is
*symmetric*.
- We express distances as dot-products (using a proof from Schonberg). This means that $ D $ is expressed as $ X^T X $. (Observe that $ X^T X $ is a matrix of dot-products).
- Once we have $ X^T X $, dimension reduction is trivial. Running an eigendecomposition on this matrix will produced centered coordinates. The low-dimension embedding is recovered by discarding eigenvalues (eigenvectors).

Thus, assuming that we work with *euclidean distances* between points, we retrieve an embedding that PCA itself would produce. Thus, **MDS with euclidean distances is identical to PCA**.

Then what exactly is the value of running MDS on a dataset?

First, the PCA is not the most powerful approach. For certain datasets, euclidean distances do not capture the shape of the underlying manifold. Running the steps of the MDS on a different distance matrix (at least one that doesn’t contain euclidean distances) can lead to better results - a technique that the Isomap algorithm exploits.

Second, the PCA requires a vector-representation for points. In several situations, the objects in the dataset are not points in a metric space (like strings). We can retrieve distances between objects (say edit-distance for strings) and then obtain a vector-representation for the objects using MDS.

In the next blog post, I will describe and implement the Isomap algorithm that leverages the ideas in the MDS strategy. Isomap constructs a distance matrix that attempts to do a better job at recovering the underlying manifold.

**PROOFS:**

Representation learning is a hot area in machine learning. In natural language processing (for example), learning long vectors for words has proven quite effective on several tasks. Often, these representations have several hundred dimensions. To perform a qualitative analysis of the learned representations, it helps to visualize them. Thus, we need a principled approach to drop from this high dimensional space to a lower dimensional space (like $ \mathbb{R}^2 $ for instance).

In this blog post, I will discuss the Multidimensional scaling (MDS) algorithm - a manifold learning algorithm that recovers (or at least tries to recover) the underlying manifold that contains the data.

MDS aims to find a configuration in a lower-dimension space that preserves distances between points in the data. In this post, I will provide intuition for this algorithm and an implementation for clojure (incanter).

One of my personal heroes, Uppalapu “Mandolin” Srinivas, breathed his last this past week. This particular album by him was my favorite:

For over five centuries the Sistine Chapel ceiling has been among the greatest things man has produced. I would give two limbs for a magnum opus of its caliber.

In contrast, the first time I saw a Rothko in my early teens, I concluded that this was the outcome of giving a child with severe OCD a set of crayons.

Over the last few weeks, amidst a very tough and frustrating period (this is far too complex for this one post) in my life, I had a chance to reflect on one of Rothko’s signature pieces and study the underlying process through a MOMA video [1]. I felt a new sense of respect for Rothko’s works. A Rothko is quite literally a metaphor for life. Our visible exterior is the product of several layers that comprise our experiences.

Rothko had a famous quote:

The people who weep before my pictures are having the same religious experience I had when I painted them.

It took eight years for me to have this experience. I am better off for it.

[1] *The Painting Techniques of Mark Rothko: No. 16 (Red, Brown, and Black)*

It is the desperate moment when we discover that this empire which had seemed to us the sum of all wonders, is an endless, formless ruin, that corruption’s gangrene has spread too far to be healed by our scepter, that the triumph over enemy sovereigns has made us the heirs of their long undoing.

— Invisible Cities (Italo Calvino)

I spent the last few months studying and implementing some routines that take a raw HTML document (or documents) and do stuff with it (them). Subotai is a library that consolidates some of these routines. In this blog post I will describe what is currently implemented and what the roadmap is.