About

Hi my name's Anthony and I'm a computer scientist. I have spent most of the last 30-odd years either building or studying the statics and dynamics of complex computational systems.

Researchwise, I am interested in algorithms that smoothly cross into the real world, like merge sort, which you can use to sort a deck of playing cards or rows in a digital database. My special interest in coevolutionary algorithmscoevolutionary algorithms
A part of complexity science that deals directly with nature-inspired evolutionary processes involving interaction in the fitness function. A definition I've used in the past is that these algorith... is motivated by my guiding belief that organisms in an ecosystem collectively embody the structure of a directed space they are simultaneously navigating as they continuously throw themselves into deep uncertainty. I enjoy exploring ways of unpacking and operationalizing this lodestar.

To that end, I've spent a lot of time thinking about latent spatial structures present in interactive domainsinteractive domains
A collection of one or more functions, called metrics, of the form $p\colon X_1\times X_2\times\cdots\times X_n\rightarrow R$, where

each $i$ with $1\leq i\leq n$ is a domain role
an element $... (and their close cousins, normal-form games) and how this structure manifests in the dynamics of algorithms endowed with informativeness incentives. To a first approximation, an algorithm such as a coevolutionary algorithm traversing such a domain with the incentive not just to progress, but also to expose that it is progressing, will tend to take states that embody this spatial structure. I named this phenomenon emergent geometric organization in my Ph.D. dissertation. I can make these notions precise for certain special cases of coevolutionary algorithm, but I think of them as broadly-applicable concepts that are also present in inchoate form in other algorithms. Old school computer science algorithms such as John Hopcroft's DFA minimization algorithm and Dana Angluin's $L^*$ algorithm; Nils Barricelli's simulated symbiogenesis; and Arthur Samuels' algorithm for learning checkers by self play, are examples.

I received a B.S. in mathematics from Case Western Reserve University in Cleveland Ohio, where I was fortunate enough to take classes with both Charles Wells and Colin McLarty. My long-running interest in category theory started with them. I completed a Ph.D. in comptuer science at Brandeis University in Waltham Massachusetts, in the DEMO lab directed by Jordan Pollack. It was at Brandeis that I first encountered coevolutionary and co-optimization algorithms, which I continue to research today. Here are some links about my research activities: