Workshop content (slides, challenges/puzzles) is also available on the [website] (https://goo.gl/Nh8HQD)
One of the major aims of this workshop is to build intuition and vocabulary which is commonly used in data analysis in most fields of biology. Another is to keep it very informal so that there is no barrier in attending the workshop. In fact, we plan to start it from scratch so that anyone can join in without hesitation. Because the basics topics covered here are identical to the ones used in machine learning, the workshop will be done under the aegis of RUSMALAI.
Although there will be only one session per week, we would be happy to discuss throughout the week (preferably post-work hours). We plan to give optional things to think about and would be glad to have a discussion on these or any other thing related to the topics going on. For such chit-chat/discussions please do walk in to the Simons Center ground floor or Sane lab.
The idea is to start with basic concepts like probability and linear algebra, and build slowly to more advanced topics like Bayesian inference, Markov processes, PCA, etc. Below is the plan for the workshop (will be updated incrementally).
Outline/breakdown for the module on Probability & Statistics:
-
Day 1 (15 February 2018): Revisiting the basics
- The premise of probability
- Writing numbers for words for simple problems: coins, dice, cards
- Conditionality, Dependence and Independence of events
- Understanding some examples from biology
-
Day 2 (22 February 2018): Distribution functions and how they come about
- Examples of discrete variables (probability mass function)
- Examples of continuous variables (probability distribution functions)
- Benefits and caveats of using Cumulative Distribution Functions (CDFs)
-
Day 3 (8 March 2018): Law of large numbers and Statistical tests
- Understanding the Normal (Gaussian) Distribution
- The basic idea of statistical tests
- Simple parametric statistical tests
-
Day 4: Bayesian approach to probability
- Bayesian method with examples
- posterior/prior distribution with examples
- Examples and implementation of Bayesian methods
- Bootstrap methods and highlight some problems
-
Day 5: Maximum Likelihood Estimation, Maximum Aposteriori Probability and practical uses of them
In planning...