sampling   3051

« earlier    

Opinion | You Don’t Want a Child Prodigy - The New York Times
"One Thursday in January, I hit “send” on the last round of edits for a new book about how society undervalues generalists — people who cultivate broad interests, zigzag in their careers and delay picking an area of expertise. Later that night, my wife started having intermittent contractions. By Sunday, I was wheeling my son’s bassinet down a hospital hallway toward a volunteer harpist, fantasizing about a music career launched in the maternity ward.

A friend had been teasing me for months about whether, as a parent, I would be able to listen to my own advice, or whether I would be a “do as I write, not as I do” dad, telling everyone else to slow down while I hustle to mold a baby genius. That’s right, I told him, sharing all of this research is part of my plan to sabotage the competition while secretly raising the Tiger Woods of blockchain (or perhaps the harp).

I do find the Tiger Woods story incredibly compelling; there is a reason it may be the most famous tale of development ever. Even if you don’t know the details, you’ve probably absorbed the gist.

Woods was 7 months old when his father gave him a putter, which he dragged around in his circular baby-walker. At 2, he showed off his drive on national television. By 21, he was the best golfer in the world. There were, to be sure, personal and professional bumps along the way, but in April he became the second-oldest player ever to win the Masters. Woods’s tale spawned an early-specialization industry.

And yet, I knew that his path was not the only way to the top.

Consider Roger Federer. Just a year before Woods won this most recent Masters, Federer, at 36, became the oldest tennis player ever to be ranked No. 1 in the world. But as a child, Federer was not solely focused on tennis. He dabbled in skiing, wrestling, swimming, skateboarding and squash. He played basketball, handball, tennis, table tennis and soccer (and badminton over his neighbor’s fence). Federer later credited the variety of sports with developing his athleticism and coordination.

While Tiger’s story is much better known, when sports scientists study top athletes, they find that the Roger pattern is the standard. Athletes who go on to become elite usually have a “sampling period.” They try a variety of sports, gain a breadth of general skills, learn about their own abilities and proclivities, and delay specializing until later than their peers who plateau at lower levels. The way to develop the best 20-year-old athlete, it turns out, is not the same as the way to make the best 10-year-old athlete.

The same general pattern tends to hold true for music, another domain where the annals of young prodigies are filled with tales of eight hours of violin, and only violin, a day. In online forums, well-meaning parents agonize over what instrument to pick for a child, because she is too young to pick for herself and will fall irredeemably behind if she waits. But studies on the development of musicians have found that, like athletes, the most promising often have a period of sampling and lightly structured play before finding the instrument and genre that suits them.

In fact, a cast of little-known generalists helped create some of the most famous music in history. The 18th-century orchestra that powered Vivaldi’s groundbreaking use of virtuoso soloists was composed largely of the orphaned daughters of Venice’s sex industry. The “figlie del coro,” as the musicians were known, became some of the best performers in the world. The most striking aspect of their development was that they learned an extraordinary number of different instruments.

This pattern extends beyond music and sports. Students who have to specialize earlier in their education — picking a pre-med or law track while still in high school — have higher earnings than their generalist peers at first, according to one economist’s research in several countries. But the later-specializing peers soon caught up. In sowing their wild intellectual oats, they got a better idea of what they could do and what they wanted to do. The early specializers, meanwhile, more often quit their career tracks.

I found the Roger pattern — not the Tiger (or Tiger Mother) pattern — in most domains I examined. Professional breadth paid off, from the creation of comic books (a creator’s years of experience did not predict performance, but the number of different genres the creator had worked in did) to technological innovation (the most successful inventors were those who had worked in a large number of the federal Patent and Trademark Office’s different technological classifications).

A study of scientists found that those who were nationally recognized were more likely to have avocations — playing music, woodworking, writing — than typical scientists, and that Nobel laureates were more likely still.

My favorite example of a generalist inventor is Gunpei Yokoi, who designed the Game Boy. Yokoi didn’t do as well on electronics exams as his friends, so he joined Nintendo as a machine maintenance worker when it was still a playing card company before going on to lead the creation of a toy and game operation. His philosophy, “lateral thinking with withered technology,” was predicated on dabbling in many different types of older, well-understood (or “withered”) technology, and combining them in new ways, hence the Game Boy’s thoroughly dated tech specs.

Roger stories abound. And yet, we (and I include myself) have a collective complex about sampling, zigzagging and swerving from (or simply not having) ironclad long-term plans. We are obsessed with narrow focus, head starts and precocity.

A few years ago, I was invited to speak to a small group of military veterans who had been given scholarships by the Pat Tillman Foundation to aid with new careers. I talked a bit about research on late specializers and was struck by the reception, as if the session had been cathartic.

One attendee emailed me afterward: “We are all transitioning from one career to another. Several of us got together after you had left and discussed how relieved we were to have heard you speak.” He was a former member of the Navy SEALs with an undergrad degree in history and geophysics and was pursuing grad degrees in business and public administration from Dartmouth and Harvard. I couldn’t help but chuckle that he had been made to feel behind.

Oliver Smithies would have made that veteran feel better too, I think. Smithies was a Nobel laureate scientist whom I interviewed in 2016, shortly before he died at 91. Smithies could not resist “picking up anything” to experiment with, a habit his colleagues noticed. Rather than throw out old or damaged equipment, they would leave it for him, with the label “Nbgbokfo”: “No bloody good but O.K. for Oliver.”

He veered across scientific disciplines — in his 50s, he took a sabbatical two floors away from his lab to learn a new discipline, in which he then did his Nobel work; he told me he published his most important paper when he was 60. His breakthroughs, he said, always came during what he called “Saturday morning experiments.” Nobody was around, and he could just play. “On Saturday,” he said, “you don’t have to be completely rational.”

I did have fleeting thoughts of a 1-day-old harp prodigy. I’ll admit it. But I know that what I really want to do is give my son a “Saturday experiment” kind of childhood: opportunities to try many things and help figuring out what he actually likes and is good at. For now, I’m content to help him learn that neither musical instruments nor sports equipment are for eating.

That said, just as I don’t plan to push specialization on him, I also don’t mean to suggest that parents should flip to the other extreme and start force-feeding diversification.

If of his own accord our son chooses to specialize early, fine. Both Mozart and Woods’s fathers began coaching their sons in response to the child’s display of interest and prowess, not the reverse. As Tiger Woods noted in 2000: “To this day, my dad has never asked me to go play golf. I ask him. It’s the child’s desire to play that matters, not the parent’s desire to have the child play.”

On the strength of what I’ve learned, I think I’ll find it easy to stick to my guns as a Roger father."
davidepstein  children  parenting  ports  talent  2019  burnout  generalists  specialization  specialists  prodigies  rogerfederer  tigerwoods  music  performance  gunpeiyokoi  gameboy  nintendo  oliversmithies  genius  science  learning  mozart  sampling  quitting  precocity  headstarts  education  focus 
6 weeks ago by robertogreco
[1902.04707] Sampling networks by nodal attributes
"In a social network individuals or nodes connect to other nodes by choosing one of the channels of communication at a time to re-establish the existing social links. Since available data sets are usually restricted to a limited number of channels or layers, these autonomous decision making processes by the nodes constitute the sampling of a multiplex network leading to just one (though very important) example of sampling bias caused by the behavior of the nodes. We develop a general setting to get insight and understand the class of network sampling models, where the probability of sampling a link in the original network depends on the attributes h of its adjacent nodes. Assuming that the nodal attributes are independently drawn from an arbitrary distribution ρ(h) and that the sampling probability r(hi,hj) for a link ij of nodal attributes hi and hj is also arbitrary, we derive exact analytic expressions of the sampled network for such network characteristics as the degree distribution, degree correlation, and clustering spectrum. The properties of the sampled network turn out to be sums of quantities for the original network topology weighted by the factors stemming from the sampling. Based on our analysis, we find that the sampled network may have sampling-induced network properties that are absent in the original network, which implies the potential risk of a naive generalization of the results of the sample to the entire original network. We also consider the case, when neighboring nodes have correlated attributes to show how to generalize our formalism for such sampling bias and we get good agreement between the analytic results and the numerical simulations."
to:NB  network_sampling  network_data_analysis  statistics  sampling  to_teach:baby-nets 
7 weeks ago by cshalizi
WhoSampled
Discover music through samples, cover songs and remixes. Dig deeper into music by exploring direct connections among songs and artists
audio  sample  samples  sampling  music 
8 weeks ago by dv8godd

« earlier    

related tags

'king  "my  1980s  2019  80s  90s  ableton  algorithm  algorithms  amateur  ambient  amenbrother  analogheat  analogtodigital  analysis  anonymization  appropriation  archive  audio  back"  batching  beamforming  beat  beatbox  bees  bias  big-data  bilinear  biodiversity  biological_recording  biophilia  bitrate  black  blackness  blue-noise  blur  bookmarks_bar  breakbeat  burnout  calls  cassette  children  city  clouds  computation  continues  control's  convergence  copyright  creativity  culture  curation  dada  data-science  data  database  davidepstein  daw  debate  deep-learning  deletion  digitakt  digitaltoanalog  distribution  diversity  documentary  download  dpp  drumandbass  drums  education  elektron  ep  erasure  everyday  experience  experimental-design  favorites  fieldwork  filters  flowchart  flstudio  fluxus  focus  free  freeculture  funkydrummer  future  futurism  gameboy  gems  generalists  generative  genius  gimp  girls  github  glitch  goldenratio  graph  graphics  gunpeiyokoi  headstarts  hiphop  history  hn  hollywoodreporter  howto  humor  hymenoptera  hypothesis-testing  image  imbalanced  industry  insects  inspiration  intellectual-property  jazz  jittered  k-dpp  khia  language-model  language  law  learning  legal  libraries  library  listening  machine-learning  machinelearning  math  mattpharr  midi  minimalism  mix  mixer  models-and-modes  models  monitoring  montecarlo  mozart  mp3  music-business  music  musik-unterricht  musiqueconcrete  my  nature  neck  network_data_analysis  network_sampling  nintendo  noise  normal_distribution  nudge-targets  ny'  octatrack  octatrak  of  oliversmithies  online-learning  open-source  optimization  out  over  parenting  path_tracing  pee  performance  physics  pianos  platform  plunderphonics  poisson-disc  pop  ports  postgresql  precocity  probability-theory  processing  prodigies  programming  public-domain  python  quality  quitting  random  rap  rave  ray_tracing  reggie.watts  remix  rendering  representation  reservoir  resize  rogerfederer  sample  samplepoints  sampler  samplers  samples  saproxylic  science  scikitlearn  sgd  shaders  shadertoy  shading  shuffling  signals  siliconvalley  simulation  size  sky  slides  socialmedia  sound.design  sound  sounds  specialists  specialization  spitfire  spotify  standard_distribution  statistics  stewardship  struggle  studio.stuff  subsampling  surveillance  symbolism  synthesizers  takedown  talent  tapes  tech  techniques  techno  ted  texture  the  thought  tigerwoods  tiled  to-understand  to-write-about  to:nb  to_teach:baby-nets  training  triangle  tutorial  tutotials  video  vinyl  volumetric  web  weight  whiteness  | 

Copy this bookmark:



description:


tags: