printer friendly pageSTATS 298: Industrial Research for Statisticians
Masters-level research as in 299, but with the approval and supervision of a faculty adviser, it must be conducted for an off-campus employer. Students must submit a written final report upon completion of the internship in order to receive credit. Repeatable for credit. Prerequisite: enrollment in Statistics M.S. program. IMPORTANT: F-1 international students enrolled in this CPT course cannot start working without first obtaining a CPT-endorsed I-20 from Bechtel International Center (enrolling in the CPT course alone is insufficient to meet federal immigration regulations).
Terms: Aut, Win, Spr, Sum
| Units: 1
| Repeatable
3 times
(up to 3 units total)
STATS 299: Independent Study
For Statistics M.S. students only. Reading or research program under the supervision of a Statistics faculty member. May be repeated for credit.
Terms: Aut, Win, Spr, Sum
| Units: 1-5
| Repeatable
for credit
Instructors:
Baiocchi, M. (PI)
;
Candes, E. (PI)
;
Chatterjee, S. (PI)
...
more instructors for STATS 299 »
Instructors:
Baiocchi, M. (PI)
;
Candes, E. (PI)
;
Chatterjee, S. (PI)
;
Duchi, J. (PI)
;
Hastie, T. (PI)
;
Johnstone, I. (PI)
;
Liang, P. (PI)
;
Linderman, S. (PI)
;
Narasimhan, B. (PI)
;
Romano, J. (PI)
;
Sabatti, C. (PI)
;
Taylor, J. (PI)
;
Tian, L. (PI)
;
Tibshirani, R. (PI)
;
Wager, S. (PI)
;
Walther, G. (PI)
;
Wong, W. (PI)
STATS 300A: Theory of Statistics I
Finite sample optimality of statistical procedures; Decision theory: loss, risk, admissibility; Principles of data reduction: sufficiency, ancillarity, completeness; Statistical models: exponential families, group families, nonparametric families; Point estimation: optimal unbiased and equivariant estimation, Bayes estimation, minimax estimation; Hypothesis testing and confidence intervals: uniformly most powerful tests, uniformly most accurate confidence intervals, optimal unbiased and invariant tests. Prerequisites: Real analysis, introductory probability (at the level of
STATS 116), and introductory statistics.
Terms: Aut
| Units: 3
STATS 303: Statistics Faculty Research Presentations
For Statistics first and second year PhD students only. Discussion of statistics topics and research areas; consultation with PhD advisors.
Terms: Aut
| Units: 1
| Repeatable
2 times
(up to 2 units total)
Instructors:
Taylor, J. (PI)
STATS 305A: Applied Statistics I
Statistics of real valued responses. Review of multivariate normal distribution theory. Univariate regression. Multiple regression. Constructing features from predictors. Geometry and algebra of least squares: subspaces, projections, normal equations, orthogonality, rank deficiency, Gauss-Markov. Gram-Schmidt, the QR decomposition and the SVD. Interpreting coefficients. Collinearity. Dependence and heteroscedasticity. Fits and the hat matrix. Model diagnostics. Model selection, Cp/AIC and crossvalidation, stepwise, lasso. Multiple comparisons. ANOVA, fixed and random effects. Use of bootstrap and permutations. Emphasis on problem sets involving substantive computations with data sets. Prerequisites: consent of instructor, 116, 200, applied statistics course,
CS 106A,
MATH 114.
Terms: Aut
| Units: 3
STATS 310A: Theory of Probability I (MATH 230A)
Mathematical tools: sigma algebras, measure theory, connections between coin tossing and Lebesgue measure, basic convergence theorems. Probability: independence, Borel-Cantelli lemmas, almost sure and Lp convergence, weak and strong laws of large numbers. Large deviations. Weak convergence; central limit theorems; Poisson convergence; Stein's method. Prerequisites:
STATS 116,
MATH 171.
Terms: Aut
| Units: 3
STATS 311: Information Theory and Statistics (EE 377)
Information theoretic techniques in probability and statistics. Fano, Assouad,nand Le Cam methods for optimality guarantees in estimation. Large deviationsnand concentration inequalities (Sanov's theorem, hypothesis testing, thenentropy method, concentration of measure). Approximation of (Bayes) optimalnprocedures, surrogate risks, f-divergences. Penalized estimators and minimumndescription length. Online game playing, gambling, no-regret learning. Prerequisites:
EE 276 (or equivalent) or
STATS 300A.
Terms: Aut
| Units: 3
STATS 319: Literature of Statistics
Literature study of topics in statistics and probability culminating in oral and written reports. May be repeated for credit.
Terms: Aut, Win, Spr
| Units: 1
| Repeatable
for credit
STATS 323: Sequential Analysis (STATS 223)
This course will survey the history of sequential analysis from its origin in the 1940s via its continuing role in clinical trials to current activity in machine learning. Subject to the limitations of time, the following topics will be discussed: parametric and semi-parametric hypothesis testing from Wald to sequential clinical trials; fixed precision estimation; change-point detection and estimation; iterative stochastic algorithms and machine learning; anytime-valid inference; optimal stopping, dynamic programming, and stochastic control; multi-armed bandits; applications. Prerequisites: for 223,
Stats 200 or equivalent; for 323,
Stats 300A and 310A.
Terms: Aut
| Units: 3
Instructors:
Siegmund, D. (PI)
;
Paul, D. (TA)
STATS 335: The Challenge Problems Paradigm in Empirical Machine Learning and Beyond
In many fields of science and technology, empirical research has been making rapid progress by implicitly following a little-studied research paradigm (CPP) with several distinctive features: a shared public database, a common task, (for example, prediction of class labels or a response variable from given input features), an objective scoring rule that quantifies performance on that task, a leaderboard that tracks performance of submissions, and a set of enrolled competitors who each try to improve the current best-known performance on that task. In the context of Empirical Machine Learning, this is explicitly the famous "Kaggle" model; however, Kaggle didn't originate this approach, and many research disciplines follow the same ingredients, in many cases implicitly or tacitly. As we know, the CPP anchored recent claims of progress in image understanding and in natural language processing. In this course we will review the many instances and variations on the CPP that exist in modern
more »
In many fields of science and technology, empirical research has been making rapid progress by implicitly following a little-studied research paradigm (CPP) with several distinctive features: a shared public database, a common task, (for example, prediction of class labels or a response variable from given input features), an objective scoring rule that quantifies performance on that task, a leaderboard that tracks performance of submissions, and a set of enrolled competitors who each try to improve the current best-known performance on that task. In the context of Empirical Machine Learning, this is explicitly the famous "Kaggle" model; however, Kaggle didn't originate this approach, and many research disciplines follow the same ingredients, in many cases implicitly or tacitly. As we know, the CPP anchored recent claims of progress in image understanding and in natural language processing. In this course we will review the many instances and variations on the CPP that exist in modern research, including not only in the standard areas of empirical machine learning (computer vision and natural language understanding) but also in academic empirical finance and computational hard sciences. We will discuss evidence that the CPP itself is a kind of secret sauce, rather than the specific technologies that are spotlighted because of CPP. We will discuss software platforms implementing CPP, including Kaggle, but also academic platforms like CodaLab, which is often used for challenge problems in natural language processing, and Nightingale Open Science which is used for challenge problems involving potentially protected health information. Prerequisite: an introductory statistics or machine learning course.
Terms: Aut
| Units: 3
Instructors:
Donoho, D. (PI)
Filter Results: