I read random papers once in a while from the AMS Math Reviews program, and I read one recently about an MCMC approach to X-ray imaging. It was a fun, detailed look at a few different ways to do sampling, and use effective sample size to figure out which worked better when.
It did also leave me wondering what the giant X-ray machines buried 1,000 feet underground are for, though.
This turned out to be a bit of a downer, but it was a good learning exercise, and the general approach will be useful for generating test data on a different project. See notebook here.
This was about 2 hours of fuss that I wish I had avoided. With my updated Jupyter Notebooks, I need to be explicit about what conda environment for python I am using.
It is all laid out clearly, if only I had been looking in this bit of the IPython docs:
For example, using conda environments, install a
Python (myenv) Kernel in a first environment:
source activate myenv
python -m ipykernel install --user --name myenv --display-name "Python (myenv)"
And in a second environment, after making sure ipykernel is installed in it:
source activate other-env
python -m ipykernel install --user --name other-env --display-name "Python (other-env)"
Just because I missed posting for the last year, doesn’t mean I have not been writing. Perhaps I have been writing more. Here is something that I just wrote for a perspective on opportunities for machine learning in population health.
Machine learning (ML) is emerging as a technology, climbing the “peak of inflated expectations” or perhaps even starting to slip into the “trough of disillusionment”, in the terms of the technology hype cycle,[ref] and offers both opportunities and threats to population health. ML is a technique for constructing computer algorithms, and what distinguishes ML methods from other computer solutions is that, while the structure of the computer program may be fixed, the details are learned from data. This data-driven approach is now dominant in Artificial Intelligence (AI), especially through deep neural networks, and stands in contrast to the old way, an expert-algorithms approach in which rules summarizing expert knowledge were painstakingly constructed by engineers and domain specialists. ML has succeeded by trading experts and programmers for data and nonparametric statistical models. However, the applications where ML has been successfully deployed remain limited. AI luminary Andrew Ng provides this concise heuristic: “[i]f a typical person can do a mental task with less than one second of thought, we can probably automate it using AI either now or in the near future.”[ref]
The editor only wants 1,000 words, so this is getting cut.
There is a new sort of publication model in Global Health, and I just gave it a try (editor’s note: just = in the last year). Gates Open Research offers “author-led publication and open peer review”, and is available now to Gates Foundation-funded researchers.
I used it to publish a Data Note, which is a dataset together with a short description of what is in the dataset.
Have a look here: https://gatesopenresearch.org/articles/2-18/v1
Just before that year of not writing anything here, I mentioned that I have a new microsimulation platform, and it is called Vivarium. That is still true, and now it even has some documentation: https://vivarium.readthedocs.io/en/latest/
It has been the thing that kept me too busy to blog for the last year. But it did generate some aesthetically pleasing figures for test purposes, as well as some population health results of interest. More details to come.