latest news

Distractions and amusements, with a sandwich and coffee.

Without an after or a when.
•
• can you hear the rain?
• more quotes

Words are easy, like the wind;

Faithful friends are hard to find.

—William Shakespeare

These are names generated from the US Census list of names using a char-rnn recurrent neural network.

The names generated by the network appear neither in the list of names nor in a 479,000 list of English words. The names may be words or names in another language, however.

Your friends discouraged you from naming your first daughter "Ginavietta Xilly Anganelel" but you didn't listen. When you named your second daughter "Nabule Yama Janda" everyone wanted to know what your secret to having such successful children was.

Below are the alphabetically first 3–10 letter female unnames for each letter. In some cases, no names of a given length were generated for a given letter.

Cac

Cau

Daa

Deu

Edz

Ele

Fea

Fri

Hea

Hhi

Ied

Ien

Jea

Kau

Kec

Ldo

Leb

Maz

Mec

Nie

Nin

Oro

Ota

Reu

Ric

Seu

Sia

Tix

Tuu

Uan

Uid

Vad

Vea

Wie

Wil

Xai

Xon

Yka

Yra

Abal

Bbhy

Beli

Cani

Caro

Daee

Dayn

Eann

Ebha

Gori

Guda

Hael

Hari

Idek

Idla

Joga

Joon

Kace

Kade

Laan

Laha

Mada

Madi

Nado

Nian

Olce

Olly

Phry

Pide

Qisy

Qoly

Rari

Rary

Sabi

Saes

Tany

Tary

Ucte

Uida

Vatt

Vean

Wada

Waid

Xiem

Yama

Yamn

Zibi

Ziun

Adyla

Babyl

Balbo

Caccy

Cadda

Dadel

Dader

Eddra

Ededr

Fibee

Fleei

Gelan

Guita

Hacie

Haela

Idale

Idena

Janda

Jerly

Kaaly

Kacee

Laala

Lacee

Maala

Mabie

Nayle

Nelli

Olako

Olise

Payna

Phaci

Qinsy

Qolee

Raane

Racey

Sacti

Sacul

Tabbr

Tabuy

Uanga

Uayda

Vaale

Vadia

Wagwa

Walyg

Xilly

Xiwda

Yabuy

Yahye

Alelee

Babela

Bajbie

Caccye

Cacell

Dadela

Dakith

Edalla

Edelah

Feliey

Felike

Garlee

Geldie

Haishe

Haline

Idelig

Idelle

Jaccey

Jatqie

Kaceey

Kacele

Laceie

Ladira

Maarae

Maarla

Nabule

Nadera

Olchee

Olisha

Pamber

Parell

Qoesha

Qoleen

Rabina

Rabymi

Sachie

Sacola

Tafbie

Tamima

Ulieta

Ullena

Vadeta

Vandie

Waghel

Wandie

Xaique

Xillia

Yaketo

Yameka

Alissea

Barelah

Barmeta

Cacalla

Caccayc

Dalecee

Dalleer

Ebeccii

Eeenera

Farleen

Ferreda

Ganalel

Griagne

Harlean

Hayceda

Iellina

Ienetka

Jadquen

Jaqquil

Kaariko

Kabjine

Labelle

Labrice

Maadite

Maadrae

Nachlee

Naqoena

Ollisha

Oralore

Panelte

Paricel

Qilonga

Qlianna

Rabette

Racelie

Sacelie

Sacelle

Tamarie

Tamarke

Ualacie

Uibelle

Valerte

Vanelte

Waylena

Wazlein

Yadalie

Yakkina

Allalera

Bamberah

Battynkb

Caccelle

Cacellen

Dacheele

Dameline

Eetenere

Eethelie

Feairice

Gaannele

Gelneria

Hacylone

Hecticie

Iachelie

Ilabetth

Jacquine

Jaqqueyn

Kabrenee

Kacalyne

Laloytha

Langella

Maarmila

Mabylere

Nadalena

Nadmelle

Orotenne

Parleeta

Parmicia

Quettine

Rachilde

Racierda

Saaleych

Saccelle

Tasharia

Tathrika

Uuguetta

Uussuida

Valtonda

Vassicha

Wapreida

Willenee

Yaumette

Yeholaki

Bathueyna

Bealyakha

Caccalren

Caleniqsa

Dalerisha

Dannerele

Eferwrace

Elaberosh

Genelnice

Helmarita

Hemaricia

Ieanerise

Ilbebette

Jatquelyn

Kacalenne

Kacelynen

Lasheudde

Laverethe

Macarelze

Macbalica

Nompterla

Porpencia

Ramancina

Rarashera

Saccellne

Sanelline

Thashinda

Tizkiqhie

Ususuista

Uussautti

Velletita

Vellotina

Ccarleetta

Deliqheeda

Elatoresha

Elisamerie

Ginavietta

Iimameline

Ilollinina

Karestanet

Kariamarie

Lelagrelie

Lelerateta

Maccelline

Maceannica

Retaqyelle

Saraquetta

Shelolesne

Cclarleette

Elisazetlie

Elisebethle

Ikekzikeina

Ilizeblelle

Kimbhrresty

Lichiabetta

Liebetreide

Mamiammalan

Marianceran

Sherleenene

Sisselletta

You name your first child "Babton Laarco Tabrit". You name your second "Ferandulde Hommanloco Kictortick". Both see infinite success in life and you wonder why you haven't discovered neural networks sooner.

Below are the alphabetically first 3–10 letter male unnames for each letter. In some cases, no names of a given length were generated for a given letter.

Aan

Bil

Bre

Cas

Ces

Daa

Dax

Ede

Eey

Har

Hhe

Ial

Iir

Jac

Jal

Kel

Kib

Lal

Lel

Mah

Meh

Nal

Nas

Oid

Oon

Phy

Pys

Roz

Ruf

Sas

Sih

Tes

Tey

Vay

Ven

Wal

Wil

Zes

Zin

Admo

Baan

Badl

Cald

Calg

Daad

Daal

Eard

Ebax

Farn

Felb

Gaht

Gart

Haan

Haco

Iane

Idae

Jaan

Jace

Kaan

Khen

Laad

Laan

Maab

Madi

Nald

Nall

Obby

Odan

Peit

Piar

Qide

Raal

Rady

Saag

Saan

Tacy

Tany

Vaen

Vaes

Waci

Waco

Ytih

Aaton

Baane

Baart

Cabis

Cailh

Daamo

Daano

Eamon

Earis

Famry

Fandy

Gacon

Gahey

Hadel

Hagre

Idail

Idris

Jacer

Jadio

Karry

Keris

Laale

Laber

Maaro

Mabin

Naado

Naalo

Oaris

Ohale

Palio

Paric

Qebin

Qikel

Rabey

Radey

Sacon

Sadne

Tacie

Talet

Uusse

Vaeld

Valen

Wacer

Wadle

Zilal

Zloyn

Aareno

Babton

Badunt

Cadlor

Cadron

Daapis

Dabron

Earrel

Earrre

Fabery

Faicey

Gaarrh

Gadano

Haares

Habide

Ienlir

Igamar

Jaalil

Jabron

Kebitt

Kelmar

Laarco

Laarin

Maccel

Maccol

Nablan

Nacell

Ohepto

Olerrh

Paciul

Pakdon

Qicias

Qrekon

Rabwin

Raciad

Saando

Saddon

Tabrit

Tactan

Ulande

Uoseol

Vachon

Vacors

Waaren

Wabton

Xiklel

Zesian

Aarsado

Balnend

Barcick

Caduuse

Caliulo

Daalius

Daarrol

Eanondo

Earesle

Falbeus

Faloric

Ganunle

Garlard

Haameno

Habrenc

Icoolse

Idonald

Jaendie

Jajuian

Kodavio

Korgell

Laarnel

Laarrec

Maccalo

Machual

Nabtumo

Nachale

Oimolan

Ollisee

Paberto

Palducb

Quitius

Radlond

Radullo

Sacholh

Saconad

Tadrine

Tahinte

Vacelle

Vagallo

Wabbent

Wacivey

Zewrave

Aarruleu

Balibhat

Baravile

Carelcic

Carkocce

Dalevice

Danilian

Earrinto

Eberepto

Fadonocf

Farricco

Gaurlnih

Gegirald

Handerus

Harelcce

Januipan

Jarcebph

Korancin

Laleaddo

Lalenicd

Maccelce

Macchely

Nachaane

Nalaneil

Parlicco

Parreico

Randlold

Rantozer

Sachasce

Sactonae

Talentin

Tavintey

Vernilve

Vernnche

Wacellio

Waldrand

Ziliasen

Aldresdis

Barrkimad

Berganton

Carlercca

Carmencan

Darriscce

Dauguslus

Edgaronte

Eeletento

Flandinco

Flilnendy

Galrinand

Gerarmovo

Hefarordo

Helaphhey

Jeenforue

Jeffersol

Lannendan

Lanuullan

Marricice

Marridcce

Nathanaal

Oberverto

Qoaberucc

Rallisten

Rardusler

Salcieley

Salvinten

Teliberel

Tewraslel

Wiccelele

Willofvis

Atthaaneel

Brantisard

Castushart

Caucerucce

Eeverielti

Elerdrolde

Ferandulde

Flarericco

Hommanloco

Kictortick

Licoonicio

Llenelvind

Nattonanal

Oriccoomon

Rarvondard

Renaldordo

Sawvarcsas

Wengortwen

Ccrickuctof

Llantonlolm

Lunuslinzus

Micckelammy

Triddatrerd

Waldinawwan

We discuss the many ways in which analysis can be confounded when data has a large number of dimensions (variables). Collectively, these are called the "curses of dimensionality".

Some of these are unintuitive, such as the fact that the volume of the hypersphere increases and then shrinks beyond about 7 dimensions, while the volume of the hypercube always increases. This means that high-dimensional space is "mostly corners" and the distance between points increases greatly with dimension. This has consequences on correlation and classification.

Altman, N. & Krzywinski, M. (2018) Points of significance: Curse(s) of dimensionality *Nature Methods* **15**:399–400.

Inference creates a mathematical model of the datageneration process to formalize understanding or test a hypothesis about how the system behaves. Prediction aims at forecasting unobserved outcomes or future behavior. Typically we want to do both and know how biological processes work and what will happen next. Inference and ML are complementary in pointing us to biologically meaningful conclusions.

Statistics asks us to choose a model that incorporates our knowledge of the system, and ML requires us to choose a predictive algorithm by relying on its empirical capabilities. Justification for an inference model typically rests on whether we feel it adequately captures the essence of the system. The choice of pattern-learning algorithms often depends on measures of past performance in similar scenarios.

Bzdok, D., Krzywinski, M. & Altman, N. (2018) Points of Significance: Statistics vs machine learning. Nature Methods 15:233–234.

Bzdok, D., Krzywinski, M. & Altman, N. (2017) Points of Significance: Machine learning: a primer. Nature Methods 14:1119–1120.

Bzdok, D., Krzywinski, M. & Altman, N. (2017) Points of Significance: Machine learning: supervised methods. Nature Methods 15:5–6.

Celebrate `\pi` Day (March 14th) and go to brand new places. Together with Jake Lever, this year we shrink the world and play with road maps.

Streets are seamlessly streets from across the world. Finally, a halva shop on the same block!

Intriguing and personal patterns of urban development for each city appear in the Boonies, Burbs and Boutiques series.

No color—just lines. Lines from Marrakesh, Prague, Istanbul, Nice and other destinations for the mind and the heart.

The art is featured in the Pi City on the Scientific American SA Visual blog.

Check out art from previous years: 2013 `\pi` Day and 2014 `\pi` Day, 2015 `\pi` Day, 2016 `\pi` Day and 2017 `\pi` Day.

We examine two very common supervised machine learning methods: linear support vector machines (SVM) and k-nearest neighbors (kNN).

SVM is often less computationally demanding than kNN and is easier to interpret, but it can identify only a limited set of patterns. On the other hand, kNN can find very complex patterns, but its output is more challenging to interpret.

We illustrate SVM using a data set in which points fall into two categories, which are separated in SVM by a straight line "margin". SVM can be tuned using a parameter that influences the width and location of the margin, permitting points to fall within the margin or on the wrong side of the margin. We then show how kNN relaxes explicit boundary definitions, such as the straight line in SVM, and how kNN too can be tuned to create more robust classification.

Bzdok, D., Krzywinski, M. & Altman, N. (2018) Points of Significance: Machine learning: a primer. Nature Methods 15:5–6.

Bzdok, D., Krzywinski, M. & Altman, N. (2017) Points of Significance: Machine learning: a primer. Nature Methods 14:1119–1120.

In a Nature graphics blog article, I present my process behind designing the stark black-and-white Nature 10 cover.

Nature 10, 18 December 2017

In this primer, we focus on essential ML principles— a modeling strategy to let the data speak for themselves, to the extent possible.

The benefits of ML arise from its use of a large number of tuning parameters or weights, which control the algorithm’s complexity and are estimated from the data using numerical optimization. Often ML algorithms are motivated by heuristics such as models of interacting neurons or natural evolution—even if the underlying mechanism of the biological system being studied is substantially different. The utility of ML algorithms is typically assessed empirically by how well extracted patterns generalize to new observations.

We present a data scenario in which we fit to a model with 5 predictors using polynomials and show what to expect from ML when noise and sample size vary. We also demonstrate the consequences of excluding an important predictor or including a spurious one.

Bzdok, D., Krzywinski, M. & Altman, N. (2017) Points of Significance: Machine learning: a primer. Nature Methods 14:1119–1120.