Upgrade to Pro — share decks privately, control downloads, hide ads and more …

Physicists in Data Science - Statistical Physic...

szilard
April 06, 2018
67

Physicists in Data Science - Statistical Physics Day, Hungarian Academy of Sciences - April 2018 (talk in Hungarian)

szilard

April 06, 2018
Tweet

More Decks by szilard

Transcript

  1. Fizikusok az adattudományban (data science) Pafka Szilárd Chief Scientist, Epoch

    (USA) Magyar Tudományos Akadémia, Budapest 2018. április
  2. 1992- ELTE fizikus 1996-98 Monte Carlo szim., Kosterlitz-Thouless 1999- pénzügyi

    alkalmazások, Kondor Imre 2001-06 CIB Bank kockázatkezelés 2004 PhD
  3. 1992- ELTE fizikus 1996-98 Monte Carlo szim., Kosterlitz-Thouless 1999- pénzügyi

    alkalmazások, Kondor Imre 2001-06 CIB Bank kockázatkezelés 2004 PhD 2006- Kalifornia, data science 2016- CEU, UCLA (1-1 kurzus)
  4. By 2018, the US alone could face a shortage of

    140,000 to 190,000 people with deep analytical skills - McKinsey, 2011
  5. David Donoho: 50 years of Data Science GDS1: Exploratory data

    analysis and data manipulation/preparation and cleaning GDS2: Systems/databases GDS3: Programming: R GDS4: Data visualization GDS5: Modeling (statistics/machine learning) GDS6: Science of Data Science
  6. Machine Learning linear/logistic regression (early 1900s/60s) decision trees (60s/80s) neural

    networks (60s/80s) support vector machines (90s) random forests (90s) gradient boosting (90s) deep learning neural networks (2000s) Source: Hastie etal, ESL 2ed y = f(x) “Learn” f from data
  7. 10x