data-science

This paper explores inductive biases and projection functions to enhance out-of-distribution generalization in autoregressive models. With the theory in place, we aim to implement these ideas in Python and develop tests and evaluations to validate our approach.

An API for constructing and working with likelihood models; works well with algebraic.mle R package

An API for algebraic maximum likelihood fitting

An R package for backpropagation through a computational graph

In my paper, Reliability Estimation in Series Systems, I discarded a lot of research that may be interesting to pursue further. This one is about using homogeneous shape parameters for the Weibull series system, which can greatly simplify the analysis and reduce the number of parameters to estimate. It’s also quite reasonable for most well-designed systems.

Investigates maximum likelihood techniques to estimate component reliability from masked failure data in series systems. A simulation study is used to assess the sensitivity of the MLE to various parameters like sample size, masking probability, and other so on, using components with Weibull lifetimes arranged in series.

R package in narrow support of my master’s project in reliability estimation for series systems from masked failure data like the component cause of failure being unknown or the failure being censored.

An API for an algebra over distributions

An API for constructing and working with dynamic failure rates.

An API for constructing and working with likelihood models; works well with algebraic.mle R package