Posts Tagged ‘reconstructability analysis’

Reconstructed Oatmeal with Pears

October 31, 2013

Rachel Ray is famous for deconstructing familiar foodstuffs into new food experiences. She’ll take a classic sausage & rice-stuffed pepper recipe, and deconstruct it into a bed of rice with chopped pepper and sausage crumble topping, or turn a broad noodle lasagna into a narrow noodle pasta plate smothered in lasagna sauce. Recently, Sydney Oland  over at the Serious Eats website had a recipe that looked like RR had produced it herself — oatmeal pancake topped with pears and pecans. What kind of a dish might that have been deconstructed from, and is it possible to reconstruct something similar?

This is quite a departure from my usual no extra work approach to breakfast oatmeal. What fascinates me about it is that it fits right in with a mathematical/statistical technique I’ve used professionally, called Reconstructability Analysis. In RA, you take a collection of data, with many variables, and use information theory to produce a collection of simpler data sets, models, and examine which model best describes the original data. The simpler model is never as good as the original, of course, but in some cases you can perform significant simplification without losing too much predictive power. Let’s see if that will work in the kitchen.

Suppose we take our oatmeal pancake dataset and reconstruct it into a standard oatmeal dish, with embedded pears and pecans? The original recipe called for braising the pears and pecans separately, but we don’t have time for that. Besides, it would dirty another pot. Let’s just dice up a pear, or part of one, and fry it in our usual oatmeal pot. Fresh Anjou pears harvested along the Willamette yesterday and FedEx’d to our doorstep this morning are the best, but cracking a can of Bartletts will probably work just as well. We’ll compromise on a somewhat overripe Bartlett pear that we got from Safeway. The pecans can be chopped and stuck in the toaster oven for later stirrings-in. If you wanted to save on electricity, you could fry up the pears and pecans together in your oatmeal pot (that’s what I did here), add the oatmeal and liquid, give it a stir, and be on your way. Thicken the oatmeal with Wondra instead of potato flakes and you have the full pancake experience, minus the baking powder. As for the liquid, I’m not sure. Plain water might work. Apple juice wants too much of itself and tastes too tart when cooked down without sugar. Maybe adding a tablespoon of maple syrup to plain water will get the effect we’re after.

Experiment 1

Setup: 1/3 cup of stone ground rolled oats, 1/4 cup of diced pear, tablespoon of chopped pecans,  one cup of water with one tablespoon of pancake syrup, salt to taste and Wondra flour to thicken.  Fry the pears and pecans in butter in the oatmeal pot. Add the water and syrup and oatmeal, oh my, when the pears are soft, and cook for another 10 minutes or so, depending on the exact style of oats.

Results: Pretty good, in a standard oatmeal sort of way. The pear chunks cooked down to pear-sauce. The nuts seemed to give up their flavor and were just inoffensive crunchy things. The oatmeal itself was creamy/buttery tasting. It needed salt, and more syrup. I am going to try it again tomorrow, adding the pecans and pear-parts at the last minute.

Rating: *****

Experiment 2

Setup: 1/3 cup of stone ground rolled oats, 1/4 cup of diced pear, tablespoon of chopped roasted pecans,  one cup of water with two tablespoon of pancake syrup, salt to taste and potato flakes to thicken.   Add the syrup at the start, Cook for another 10 minutes or so, depending on the exact style of oats. Add the pears and nuts at the end and thicken with the flakes if you think you need it.

Results: I’ve decided I don’t like chopped nuts in my oatmeal. They don’t add anything, and they make you chew stuff that otherwise doesn’t need much chewing. Pears were more noticable this time. The residual heat (and extreme ripeness) took some of the crunch off them, but I always knew when I had some on my spoon. The lack of butter was very noticable — bad for flavor, good for calories. Overall it was OK, but not OK enough for me to spend a lot of time in the pear aisle.

Rating: *****


August 21, 2012

The 6th International Conference on Soft Computing and Intelligent Systems and the 13th International Symposium on Advanced Intelligent Systems will be held at the Kobe Convention Center in the Kobe Portopia Hotel next November. I have two papers submitted. Or, I should say, we have two papers, because in this business you don’t get anywhere without a lot of help from your friends. UPDATE: Both papers have been accepted, which is why I’m posting this here and now, after a couple of false starts.

The first paper is on the application of a Systems Science tool called Reconstructablility Analysis to understanding the genetics of Alzheimer disease. Here’s the abstract:

Reconstructability Analysis (RA) is an information- and graph-theory-based method which has been successfully used in previous genomic studies. Here we apply it to genetic (14 SNPs) and non-genetic (Education, Age, Gender) data on Alzheimer disease in a well-characterized Case/Control sample of 424 individuals. We confirm the importance of APOE as a predictor of the disease, and identify one non-genetic factor, Education, and two SNPs, one in BINI and the other in SORCS1, as likely disease predictors. SORCS1 appears to be a common risk factor for people with or without APOE. We also identify a possible interaction effect between Education and BINI. Methodologically, we introduce and use to advantage some more powerful features of RA not used in prior genomic studies.

The second paper used another Systems Science technique, agent-based simulation, to test Herb Simon’s theory of satisficing:

Satisficing is an efficient strategy for applying existing knowledge in a complex, constrained, environment. We present a set of agent-based simulations that demonstrate a higher payoff for satisficing strategies than for exploring strategies when using approximate dynamic programming methods for learning complex environments. In our constrained learning environment, satisficing agents outperformed exploring agent by approximately six percent, in terms of the number of tasks completed.

In a later post, I’ll talk about the collaboration that led up to this.

Reconstructability Analysis

July 7, 2009

Reconstructability analysis (RA) is an information- and graph-theoretic methodology which originates with Ross Ashby’s constraint analysis and was subsequently developed by several others. RA resembles log-linear methods used widely in the social sciences, and where RA and log-linear methodologies overlap they are equivalent. RA also overlaps with Bayesian networks. In RA, a probability or frequency distribution or a set-theoretic relation is decomposed into component distributions or relations. When applied to the decomposition of frequency distributions, RA does statistical analysis. RA can model problems both where “independent variables” (inputs) and “dependent variables” (outputs) are distinguished (called directed systems) and where this distinction is not made (neutral systems). Being based on information theory, which ignores metric information in the variables being analyzed, RA is a natural methodology for nominal, e.g., genomic, data.

Right now, I’m looking at how RA compares with Logistic Regression. They produce identical classification rates for low penetrance genetic data, but RA appears to be easier to use — you don’t have to create dummy variables, and the results look to be more directly readable.