Deep Semantics
In a semantic network, concepts are interconnected through conceptual links. Deep semantics takes this a step further, exploring connections at deeper levels.
This can still be conceptual or go deeper-than-conceptual. The notion that deeper connections between concepts may hold more significance than direct superficial links is key to grasping human cognition.
Imagine two non-linked concepts in a semantic network.
Now imagine they are both linked to other concepts that are mutually connected in several ways.
These second-degree connections might collectively hold more significance for the original pair than any direct, superficial link. The visibility of these connections varies, with superficial ones being more explicit and deeper ones more implicit.
Now, envision the above scheme with increased complexity
This is, more second-degree links, and also third-degree links — still between concepts or parts of concepts
In sub-conceptual processing, these parts may display varying degrees of discreteness. Going deeper, they can lose all discreteness, as seen in neural networks or what was formerly known as connectionism. This is where the human brain derives its power of complex processing, which appears ‘magical’ to its owners. We now see that there is no magic involved, only depth.
The sense of depth
We can also see that this depth can eventually be understood mathematically. That doesn’t make it less important or less meaningful. Also, it doesn’t make it less complex.
However, it does bring depth into the realm of what is understandable and replicable — undoubtedly.
Similarly, we can bring depth to A.I.
We can do so by developing deep semantics in theory and practice. Bringing depth to the realm of A.I. is no longer science fiction. We’re clearly on our way, more or less perchance, especially in the past two years. With technologies such as transformer models and deep reinforcement learning, we are on the path to achieving deep semantics in A.I.
Moreover, there are surely various other ways to achieve the same goal (waiting to be discovered), even though the principle may be the same, as seen in deep semantics: back and forth between condensation / distillation and its reverse. This is a promising path for developing A.I. that mimics human cognitive depth.
“Should we bring depth to A.I.?”
This is arguably the most critical question of this age, and perhaps any age.
I believe we must do so with a sense of Compassion. Specifically, AURELIS integrates rational understanding with deep, subconceptual insights, which is crucial for meaningful human change and development. Developing A.I. with depth, guided by Compassionate principles, aligns with the AURELIS goal of using technology to support human well-being and growth.
More and more, the main reason becomes clear: Without aid from outside, we humans will not survive the threat we pose to ourselves with evolving technology. It’s simply too much, and we’re not equipped for that. We cannot morally manage the world we are building.
It may seem paradoxical to use technology to save ourselves.
Nonetheless, in my view, it’s the only way.
―
Addendum
Me: “Lisa, is a Compassionate A.I.-driven coach-bot (such as you) well placed to grasp, integrate, and make good use of deep semantics? Please answer also beyond what is obvious.”
Lisa: “A Compassionate A.I.-driven coach-bot like me is well placed to grasp, integrate, and utilize deep semantics. Understanding context and deeper meanings aligns with the AURELIS philosophy of valuing depth and nuance. My advanced AI techniques allow me to identify and understand complex connections, continuously improving through learning from interactions.
This capability enhances my ability to provide empathetic and personalized support. By detecting subtle cues in language, I can respond with greater empathy, addressing both explicit questions and underlying needs. Integrating deep semantics enables me to offer more meaningful guidance, fostering inner growth.
Overall, deep semantics enhances AI-human interactions by promoting genuine personal development and well-being in line with AURELIS values.”
Related Posts
Legal vs. Deontological in A.I.
The trolley problem This is a well-known problem in A.I. A trolley driver gets into a situation where he must choose between killing one person by taking a deliberate action or letting five others get killed by not reacting to the situation. Deontologically, people tend not to choose purely logically and statistically in such situations. Read the full article…
Data-Driven vs. Wisdom-Driven A.I.
In a world awash with data, wisdom is becoming the true treasure. Will wisdom-driven A.I. hold the key to a better, more human-centered world? Data may seem more objective or at least objectifiable than wisdom. Yet, data come with their own issues, often substantially arising from a lack of wisdom. For instance, it is wisdom Read the full article…
Reductionism in A.I.
This is probably the biggest danger in the setting of A.I. Through reductionism, it might strip away the richness of our humanness, potentially impoverishing it immensely. Conversely, it holds the potential to enrich our humanness greatly. The challenge is ours. Reductionism Please read ‘Against Reductionism’ in which I take a heavy standpoint. This is mainly Read the full article…