Thoughts about Scientific Machine Learning and AI – Michael Green

Categories

All (22)

accountability (1)

ai (2)

bayesian (10)

bimodality (1)

cacophony (1)

causality (1)

chemistry (1)

cloud (1)

coco (1)

confidence intervals (1)

conjugacy (1)

constrained optimization (1)

corona (1)

correlation (1)

credible interval (1)

deep learning (2)

derivative (1)

differentialequations (1)

dirichlet-multinomial (1)

drug discovery (1)

econometrics (1)

elasticity (1)

epidemiology (1)

europe (1)

european (1)

evolutionary (1)

frequentist (2)

human-rights (1)

hypothesis testing (2)

inference (2)

infrastructure (1)

international-law (1)

julia (1)

learning (2)

likelihood (1)

logd (1)

logp (1)

logs (1)

marketing mix modeling (1)

math (1)

maximum likelihood (2)

ml (1)

model (1)

modeling (6)

molecular properties (1)

multinomial (1)

nextcloud (1)

non-linear (3)

object detection (1)

optimization (2)

p-value (1)

pandemic (1)

personal development (2)

pKa (1)

posterior (1)

price (1)

prior (5)

programming (1)

rant (1)

regression (8)

significance (1)

simulation (2)

statistics (1)

technology (1)

variable selection (1)

virus (1)

xi correlation (1)

A Law That Should Disturb Every One of Us

On Israel’s death penalty law targeting Palestinians, and why silence is not an option

international-law

I don’t usually write about legislation. My usual territory is neural networks, Bayesian methods, and the occasional rant about European tech sovereignty. But some things…

How I got rid of American Technology and why you should too

For those living in Europe, it’s no secret that we are navigating a complex and volatile global landscape. Recent actions by the United States including its inhumane stance…

The cacophony around us

personal development

Never before have I heard so many people speak of so many things they know so little about.

Learning a new skill: A thing of the past?

personal development

Most people assume that AI will eventually do everything. That’s wrong. AI is a tool, a powerful one, but still just a tool. Thinking AI will replace human skill is like…

Evidential musings

Utilizing conjugacy for the Multinomial distribution

dirichlet-multinomial

As outlined in Sensoy et al. (2018) there’s some work done in Evidential Deep Learning to quantify uncertainty in classification applications. The evidential framework…

ξ correlation: A new tool for your causality checks

There’s a new candidate for checking for relationships between two variables out on the streets. It’s been there since 2020, but I actually just heard about it now. It’s…

LogP, LogD, pKa and LogS: A Physicists guide to basic chemical properties

molecular properties

Ever since I got into AI Based Drug Discovery a lot of terminology has been thrown my way. The idea to write this came from (Bhal 2021) which outlines that it’s important to…

What does the prior actually do in a Bayesian analysis?

I often feel when talking to coworkers, practitioners and people in general that the Bayesian philosophy has to prove it’s worth or justification as if it’s a contender to…

COVID-19 in Denmark - An epidemic in a small country

differentialequations

The COVID-19 virus has struck the world with a rare force (Hui et al. 2020). In a matter of months it has affected 194 countries and territories so far. With respect to…

A quick introduction to derivatives for machine learning people

If you’re like me you probably have used derivatives for a huge part of your life and learned a few rules on how they work and behave without actually understanding where it…

The importance of context

maximum likelihood

When we do modeling it’s of utmost importance that we pay attention to context. Without context there is little that can be inferred.

Dr. Michael Green

Deep Neural Networks in Julia - Love at first sight?

I love new initiatives that tries to do something fresh and innovative. The relatively new language Julia is one of my favorite languages. It features a lot of good stuff in…

On the apparent success of the maximum likelihood principle

maximum likelihood

Today we will run through an important concept in statistical learning theory and modeling in general. It may come as no surprise that my point is as usual “age quod agis”.…

Dr. Michael Green

Building and testing a simple deep learning object detection application

object detection

Deep learning is hot currently. Really hot. The reason for this is that there’s more data available than ever in the space of perception. By perception I mean tasks such as…

Dr. Michael Green

About identifiability and granularity

marketing mix modeling

variable selection

In time series modeling you typically run into issues concerning complexity versus utility. What I mean by that is that there may be questions you need the answer to but are…

Dr. Michael Green

A few thoughts on apparent bimodality for regression problems!

Did you ever run into a scenario when your data is showing two distinctive relationships but you’re trying to solve for it with one regression line? This happens to me a…

Dr. Michael Green

On the equivalence of Bayesian priors and Ridge regression

Today I’m going to take you through the comparison of a Bayesian formalism for regression and compare it to Ridge regression which is a penalized version of OLS. The…

Dr. Michael Green

About choosing your optimization algorithm carefully

constrained optimization

Why is simulation important anyway? Well, first off we need it since many phenomemna (I would even say all interesting phenomemna) cannot be encapsulated by a closed form…

Dr. Michael Green

The truth about priors and overfitting

hypothesis testing

Have you ever thought about how strong a prior is compared to observed data? It’s not an entirely easy thing to conceptualize. In order to alleviate this trouble I will take…

Dr. Michael Green

The insignificance of significance

hypothesis testing

Statistical significance has always held a slightly magical status in the research community as well as in every other community. This position is unwarrented and the trust…

Dr. Michael Green

Doing price optimization in R

As many of us already know R is an extremely useful and powerful language for designing, building and evaluating statistical models. In this example I’m going to use R for…

Dr. Michael Green

About confidence intervals

confidence intervals

credible interval

Confidence intervals are as beautiful as they are deceiving. They’re part of an elegant theory of mathematical statistics which has been abused since the dawn of time. Why…

Dr. Michael Green