Gustavo Lacerda

From Santa Fe Institute Events Wiki

on vacation in Brooklyn, NY (June 2008)

About me

I'm a grad student at UBC, interested in machine learning and information theory.

Although I'm a converted pragmatist, I still like pretty math especially when it helps us answer questions about the real world. I have a knack for spotting ambiguities and paradoxes, and sometimes need to resist the temptation of fixing them (a little handwaving is sometimes necessary for fast progress). My most developed nameable skill seems to be programming (this is distinct from "hacking", at which I'm merely ok). I can be a pedanticist about notation, and often wish that standard mathematical language were based on typed lambda calculus. I'm a former semantic web dreamer, but remain fond of argument diagramming.

My background is diverse. I have degrees in Mathematics (BS) and Logic (MSc). My most important publication so far was at UAI2008 (using ICA to infer the structure of causal networks). I'm a reviewer for CogSci, and I've recently had recommendations written by profs in Computer Science, Philosophy, Statistics, and Brain and Cognitive Sciences.

Recently concluded projects include:

  • search algorithms for Bayesian sparse linear regression
  • stochastic gradient methods for training Restricted Boltzmann Machines
  • time-series modeling of energy usage in office buildings

At SFI, I'm hoping to boost my research on complex networks, especially if it relates to genomics.

The Questions

1. What are your main interests? Feel free to include a "pie in the sky" big idea!

  • making predictions from/about complex-structured data (a.k.a. "Learning with structured input and output")

2. What sorts of expertise can you bring to the group?

  • statistical modeling / machine learning / Bayesian inference
  • programming
  • artificial intelligence
  • an understanding of conceptual issues related to philosophy of science and methodology, especially causality (philosophy has been a serious hobby of mine)

3. What do you hope to get out of the CSSS?

  • ideas for a long-term project collaborations
  • be in a position to bridge gaps between different research communities

4. Do you have any possible projects in mind for the CSSS? (Recall that you will all be working in groups on at least one project with the goal of presenting your progress on the last day and finishing up a paper by sumer's end.)

  • data mining on large networks.
  • I'm currently interested in inferring causality from dynamical systems. I should study my stochastic differential equations.



email: gusa at the domain below