michael-jordan   34

Duchi, Jordan, Wainwright, "Local Privacy and Statistical Minimax Rates"
Introduces "local differential privacy," where data remains private even from the learner or data analyst.
local-differential-privacy  differential-privacy  research-article  michael-jordan  privacy 
9 days ago by arthegall
The comparison between LeBron James and Michael Jordan, in their own words
"I watch Jordan more than anybody; for sure, MJ wasn't perfect. MJ had bad games. He had turnovers. He had games where he felt like he should have been better. But I think the greatest thing about MJ was that he never was afraid to fail. And I think that's why he succeeded so much, because he was never afraid of what anybody ever said about him. Never afraid to miss a game-winning shot, never afraid to turn the ball over. Never afraid. That's one of my biggest obstacles. I'm afraid of failure. I want to succeed so bad that I become afraid of failing."
-- LeBron James
Oct. 19, 2013 (Source: ESPN)
lebron-james  michael-jordan  success  failure  mindset-success  mental-barriers 
9 weeks ago by lwhlihu
Stat 260/CS 294: Bayesian Modeling and Inference
- Priors (conjugate, noninformative, reference)
- Hierarchical models, spatial models, longitudinal models, dynamic models, survival models
- Testing
- Model choice
- Inference (importance sampling, MCMC, sequential Monte Carlo)
- Nonparametric models (Dirichlet processes, Gaussian processes, neutral-to-the-right processes, completely random measures)
- Decision theory and frequentist perspectives (complete class theorems, consistency, empirical Bayes)
- Experimental design
unit  course  berkeley  expert  michael-jordan  machine-learning  acm  bayesian  probability  stats  lecture-notes  priors-posteriors  markov  monte-carlo  frequentist  latent-variables  decision-theory  expert-experience  confidence  sampling 
july 2017 by nhaliday
How to Escape Saddle Points Efficiently – Off the convex path
A core, emerging problem in nonconvex optimization involves the escape of saddle points. While recent research has shown that gradient descent (GD) generically escapes saddle points asymptotically (see Rong Ge’s and Ben Recht’s blog posts), the critical open problem is one of efficiency — is GD able to move past saddle points quickly, or can it be slowed down significantly? How does the rate of escape scale with the ambient dimensionality? In this post, we describe our recent work with Rong Ge, Praneeth Netrapalli and Sham Kakade, that provides the first provable positive answer to the efficiency question, showing that, rather surprisingly, GD augmented with suitable perturbations escapes saddle points efficiently; indeed, in terms of rate and dimension dependence it is almost as if the saddle points aren’t there!
acmtariat  org:bleg  nibble  liner-notes  machine-learning  acm  optimization  gradient-descent  local-global  off-convex  time-complexity  random  perturbation  michael-jordan  iterative-methods  research  learning-theory  math.DS  iteration-recursion 
july 2017 by nhaliday
michaelijordan comments on AMA: Michael I Jordan
"I'm also overall happy with the rebranding associated with the usage of the term "deep learning" instead of "neural networks". In other engineering areas, the idea of using pipelines, flow diagrams and layered architectures to build complex systems is quite well entrenched, and our field should be working (inter alia) on principles for building such systems. The word "deep" just means that to me---layering (and I hope that the language eventually evolves toward such drier words...). I hope and expect to see more people developing architectures that use other kinds of modules and pipelines, not restricting themselves to layers of "neurons"."

"Indeed, it's unsupervised learning that has always been viewed as the Holy Grail; it's presumably what the brain excels at and what's really going to be needed to build real "brain-inspired computers". But here I have some trouble distinguishing the real progress from the hype. It's my understanding that in vision at least, the unsupervised learning ideas are not responsible for some of the recent results; it's the supervised training based on large data sets.
One way to approach unsupervised learning is to write down various formal characterizations of what good "features" or "representations" should look like and tie them to various assumptions that seem to be of real-world relevance. This has long been done in the neural network literature (but also far beyond). I've seen yet more work in this vein in the deep learning work and I think that that's great. But I personally think that the way to go is to put those formal characterizations into optimization functionals or Bayesian priors, and then develop procedures that explicitly try to optimize (or integrate) with respect to them. This will be hard and it's an ongoing problem to approximate. In some of the deep learning learning work that I've seen recently, there's a different tack---one uses one's favorite neural network architecture, analyses some data and says "Look, it embodies those desired characterizations without having them built in". That's the old-style neural network reasoning, where it was assumed that just because it was "neural" it embodied some kind of special sauce. That logic didn't work for me then, nor does it work for me now."
michael-jordan  deep-learning  neural-net  unsupervised 
november 2014 by arsyed

related tags

acm  acmtariat  ai  approximation  art  arxiv  assignment-chicago  bagging  basketball  bayesian  ben-recht  berkeley  beta-process  big-data  bill-simmons  blackhawks  blogpost  book  boostrap  bulls  cats  cavs  characterization  chicago  classification  commented  composition-decomposition  concept  confidence  convex-optimization  cool  course  crag-ehlo  decision-theory  deep-learning  design  differential-privacy  discriminative-training  empirical  estimate  examples  expert-experience  expert  exposition  extrema  failure  frank-chimero  frequentist  frontier  fuckyeahchicago  funny  gifs  gradient-descent  highlights  idea  information-theory  intricacy  ios  iteration-recursion  iterative-methods  james-dator  jordan  kernels  kobe-bryant  latent-variables  lda  learning-theory  learning  lebron-james  lecture-notes  lenkody  levers  linear-algebra  liner-notes  local-differential-privacy  local-global  local-privacy  machine-learning  machinelearning  markov  martin-wainwright  math.ds  memes  mental-barriers  mindset-success  mj  mobile  model-class  monte-carlo  motivation  nba  neural-net  nibble  numerics  off-convex  optimization  org:bleg  papers  parsimony  pdf  perturbation  phil-jackson  priors-posteriors  privacy  probabilistic-methods  probability  random  ranking  realness  reduction  reflection  replication  research-article  research  rigor  robust  rounding  sampling  sbnation  science  sports  sportswriting  statistics  stats  stories  story  success  summary  supreme  systematic-ad-hoc  table  talks  the-trenches  tightness  time-complexity  time-series  topic-models  tricks  tweet  unc-chapel-hill  unc-tarheels  unit  unsupervised  video  videos  vince-carter  windycitizen  youtube 

Copy this bookmark: