The nsm theory is a theory that…

1. claims that there exists a set of semantic primes and logic universal across languages which is indefinable by other words within the language
2. which, as a corollary, resolves the epistemological problem that if all words are defined by other words in the language there will (why?) be no connection to the real world

the theory of NSM rests on…

two pillars of NSM theory

existence of semantic primes

The existence of semantic primes is codified more formally as the strong version of the Lexicalization Hypothesis.

NBBO is the composite best bid/ask nationally, across all of the exchanges.

It allows the average person on the street to get a price for the asset. The government system is actually SLOWER from the fastest exchange: you can know, within microseconds, the difference.

A very simple machine learning method.

For some input \(\qty (\hat{x}, \hat{y})\), find the closest training data point \(\qty (x_{i}, y_{i})\) and return its associated \(z_{i}\), with error \(\mid z_{i} - \hat{z}\mid\). Consider a Voronoi Diagram!

This surface is piecewise constant.

see also nearest neighbor method and data interpolation

Framework for subsurface exploration: a DARPA challenge that explores unknown underground environments.

Main Problem: there is a high degree of uncertainty that comes from multiple different systems interacting:

• sensing
• environment
• command execution
• communication
• mission state
• health of systems and subsystems

NeBula treats uncertainty between systems via a POMDP:

• construct a simulation of the tasks to coordinate robots
• solved using Double Progressive Widening

AISR NeBula

NeBula autonomy framework extrapolation on an active source seeking. For instance, combining with semantic understanding, we want to “find the red backpack”.