From beef8300549b2e1a2d9a9dedf2cf7737e723efdc Mon Sep 17 00:00:00 2001
From: howthebodyworks
Date: Thu, 19 Jan 2012 18:09:10 +1100
Subject: [PATCH] refactor plans
---
TODO.rst | 15 +++++++--------
1 file changed, 7 insertions(+), 8 deletions(-)
diff --git a/TODO.rst b/TODO.rst
index 7b34d38..f335bc2 100644
--- a/TODO.rst
+++ b/TODO.rst
@@ -3,7 +3,6 @@ TODO
* Basic functionality
- * Why is the key for each experiment different?
* ditch in-process dispatcher. unneccessary complexity.
* Sweep and bootstrap param ranges
* Revise the swappable evaluator function protocol to pass all state in -
@@ -18,11 +17,7 @@ TODO
* Implement firm death?
* Implement firm replacement?
- * Simplify skill and culture values by restricting them to a smart range -
- make the skill scalar be between zero and 1, and culture vector be binary
- vector. Then the effectiveness can be based on a weighted geometric mean
- of aggregate culture vector and mean skill.
- * Groysberg's data is based on partial information?
+ * Groysberg's data is based on *partial* information?
* Actual data parameters:
* 20000 workers, 400 firms, ca. 30 timesteps (annual rankings)
@@ -33,6 +28,9 @@ TODO
unknown "correct" mean culture vector, which firms may try to approximate.
This has the potential for many local optima.
* record aggregate per-firm culture vector value in logs
+ * Base effectiveness can be based on a weighted geometric mean of aggregate
+ culture vector and mean skill. (will need a more sophisticated version of
+ marginal utility ofa single firm)
* Worker valuation
@@ -47,7 +45,7 @@ TODO
plus continuous (profitability)
* Can we, should we, get them to value workers for themselves versus
- everyone else and bid accordingly?
+ everyone else and bid accordingly, i.e. strategically?
* No firm-private information at the moment. Should this be an open-kimono
sim? How do we get asymmetrical judgements then?
* more complex schemes are alarming - we know the dimensionality of the
@@ -66,7 +64,8 @@ TODO
equation)
* raw info-theory one- we know how many bits there are to learn in each
agent, so we can bound the convergence rate of the estimator if we
- know model structure and it is mere parameter estimation
+ know model structure and it is mere parameter estimation. Can we bound
+ the estimation accuracy by considering approximation theory
* real valuations would include the strategic component of the desirability
of a worker to others (though a random bid in the tenable range is all I
--
2.1.1