Probabilistic Inference

When we consider complex concepts like friendship, we can struggle to define them definitively. Is friendship more internal or external? More objective or subjective?

Infinite precision is infinitely expensive. By converting this uncertainty into confidence levels, we can get practical, usable answers from partial certainties cheaply.

Degrees, Not Boxes

Let’s replace the binary yes-or-no with degree measurements. Instead of “Is this internal or external?” we can ask “How internal and how external is this?”

Then, the probability for a given category is simply the product of the probabilities.

\begin{aligned} P(Know) &= P(Internal) \times P(Objective) \times P(Relative) \\ P(Sense) &= P(External) \times P(Objective) \times P(Absolute) \\ P(Feel) &= P(Internal) \times P(Subjective) \times P(Absolute) \\ P(Predict) &= P(External) \times P(Subjective) \times P(Relative) \end{aligned}

Think of the friendship that you had with your best friend. It is between both of you (internal) and is based on your personal history (subjective). You understand friendship without comparison: romantic love is a different kind of relationship, just as a square is a different shape from a triangle (absolute).

• That said, you can see if someone is friends with someone, so there is some uncertainty: P(Internal) = 0.65, P(External) = 0.35
• More subjective, starts with you: P(Objective) = 0.3, P(Subjective) = 0.7
• Strongly absolute: P(Absolute) = 0.85, P(Relative) = 0.15

P(Feel) has the highest score at 0.39 (39%), followed by P(Sense) at 0.09, P(Predict) at 0.04, and P(Know) at 0.03.

You didn’t need to decide whether friendship is definitely internal. Feeling still won with 39%, with everything else following far behind. You did not need 100% certainty to make decisions that work.

That 16% attributable to other operations is what we lost to aliasing across many dimensions of friendship (a complex socio-biological phenomenon) when we fit it into one of the four boxes.

Selective Filter

If we compute the totals, we find that almost half the probability (46%) has no place in any operation. That’s more than the highest-ranking answer. It isn’t aliased-away information about friendship; it’s information that meant nothing, void.

No operation is all internal, subjective, and relative, yet this contradiction accounts for 21% of the total probability. The system excluded background noise using structural constraints between attributes.

Recall the sufficiency constraint. If the environment has infinite ways to kill you, then most structures of matter don’t persist. Life has to find the tiny subset of designs that work against perpetual chaos.

Later we will expand our classification. The more attributes we add, the more selective the filter becomes. A classification with more types and attributes that selects only 5% of meaning from the chaos implies that coherent structures are rare.

Protocells did not stop at learning four operations just to attend to the world. There’s not much to just listening, but if they can act and respond in the same four-verb language, that’s an evolutionary edge.