Coherent Aggregated Volition

From Lesswrongwiki
Revision as of 01:26, 3 October 2012 by Pedrochaves (talk | contribs)
Jump to: navigation, search

Coherent Aggregated Volition is Ben Goertzel's response to Eliezer Yudkowsky's Coherent Extrapolated Volition. CAV would be a combination of the goals and beliefs of humanity at the present time.

The author considers the "extrapolation" aspect of CEV as distorting the concept of volition and to be highly uncertain. He considers that if the person whose volition is being extrapolated has some inconsistent aspects (which is tipically human), then there could be a great variety of possible extrapolations. The problem would then be which version of this extrapolated human to choose, or how to aggregate them, which would be very difficult to achieve.

Coherent Aggregated Volition is presented as simpler than his interpretation of CEV, and intended to be easier to formalize and prototype in the foreseeable future. CAV is not, however, intended to answer the question of Friendly AI, although Goertzel claims CEV is possibly not the answer as well.

The concept

The author starts by defending that we must treat goals and beliefs together, as a single concept, which he calls gobs (and gobses for the plural). Each agent thus can have several gobs, logically consistent or not. As a way to measure how much these gobs are distant to each other, the term gobs metric is used - the persons or AGI could agree, with various degrees, on several metrics, but it seems probable that this individual's metrics would differ less than their gobses.

Then, given a population of intelligent agents with different gobses, we could then try to find a single gobs that maximizes logical consistency, compactness, similarity to the different gobses in the population and amount of evidence supporting these beliefs. This "multi-extremal optimization algorithm" is what he calls Coherent Aggregated Volition. The term expresses the attempt to achieve both coherence and an aggregation of the population volitions.

CAV has some free parameters, like the averaging method, the measure of compactness, consistency evaluation and so on, but these are seen as features rather than limitations and do not taint the simplicity of the idea. At the same time, it is possible to refine some of the criteria stated before without changing the nature of the method.


Although CEV is seen as possibly giving a feasible solution, Goertzel states there's no guarantee of this, and that Yudkowsky's method can generate solutions very far from the populations' gobses.

In some experiments with iteratively repairs within a probabilistic reasoning system, the author claims it seems we can reach a set of beliefs very different from the one we started - which seems to reflect the CEV process and main problem. That is, the iterative refinement of the agents' goals and beliefs can not always be a good system to turn inconsistent values with similar consistent ones.

Finally, while it seems that CEV bypasses the humanity feeling, by not encompassing the person's growth process and inconsistencies in the human mind, CAV tries to summarize it, respecting and not replacing it. The intrinsic inheritance of more "bad" aspects of humanity than CAV is then seen as more human.

See also