Gordon Worley

The AGI alignment problem has a bimodal distribution of outcomes with most outcomes clustering around the poles of total success and existential, catastrophic failure. Consequently, attempts to solve AGI alignment should, all else equal, prefer false negatives (ignoring research programs that would have been successful) to false positives (pursuing research programs that will unexpectedly fail). Thus, we propose adopting a policy of responding to points of philosophical and practical… Show more

The AGI alignment problem has a bimodal distribution of outcomes with most outcomes clustering around the poles of total success and existential, catastrophic failure. Consequently, attempts to solve AGI alignment should, all else equal, prefer false negatives (ignoring research programs that would have been successful) to false positives (pursuing research programs that will unexpectedly fail). Thus, we propose adopting a policy of responding to points of philosophical and practical uncertainty associated with the alignment problem by limiting and choosing necessary assumptions to reduce the risk of false positives. Herein we explore in detail two relevant points of uncertainty that AGI alignment research hinges on---meta-ethical uncertainty and uncertainty about mental phenomena---and show how to reduce false positives in response to them. Show less

Defensive alliances are a way of using graphs to model the defense of resources (people, buildings, countries, etc.) against attacks where the number of potential attackers against each resource is known. The initial study of defensive alliances focused on questions of minimal defensive alliances in a graph and the minimum possible size of a defensive alliance in a graph, but in order to apply defensive alliances in modeling real-world situations, additional considerations are important. In… Show more

Defensive alliances are a way of using graphs to model the defense of resources (people, buildings, countries, etc.) against attacks where the number of potential attackers against each resource is known. The initial study of defensive alliances focused on questions of minimal defensive alliances in a graph and the minimum possible size of a defensive alliance in a graph, but in order to apply defensive alliances in modeling real-world situations, additional considerations are important. In particular, since each vertex in a defensive alliance represents some real-world object that has a cost associated with remaining in the defensive alliance, it is important to consider the value each vertex adds to the defensive alliance. In this thesis we consider a method of assessing the efficiency of a defensive alliance, including the special case of secure sets. Show less

Given a graph G, the security number of G is the cardinality of a minimum secure set of G, the smallest set of vertices S ⊆ V (G) such that for all X ⊆ S, |N[X] ∩ S| ≥ |N[X] − S|. It is believed to be computationally difficult to find the security number of large graphs, so we present techniques for reducing the difficulty of both finding a secure set and determining bounds on the security number.

In 2003, Bechmann-Pasquinucci introduced the concept of quantum seals, a quantum analogue to wax seals used to close letters and envelopes. Since then, some improvements on the method have been found. We first review the current quantum sealing techniques, then introduce and discuss potential applications of quantum message sealing, and conclude with some discussion of the limitations of quantum seals.

We present a so-called fuzzy watermarking scheme based on the relative frequency of error in observing qubits in a dissimilar basis from the one in which they were written. Then we discuss possible attacks on the system and speculate on how to implement this watermarking scheme for particular kinds of messages (images, formated text, etc.).