Question

Is it possible to create a ‘Transverse Field Ising Spin’-compatible Super Hamiltonian?

I want to apply the Super Hamiltonian to this paper: https://arxiv.org/abs/1612.05695

Comments

What is a super Hamiltonian?

---

See this paper on Super Hamiltonian: https://arxiv.org/abs/hep-th/0506170

---

It seems like that paper has much more to do with quantizing classical mechanical systems than lattice Hamiltonian models like Ising. What do you have in mind?

---

It concerned Artificial General Intelligence.

Based on talks with physicists elsewhere, I had come to somewhat resolve the issue.

This is the outcome, based on this.

See a clear overview of the outcome here.

Edited to replace academia links with researchgate links, I deleted my academia profile on October 1'st, due to too many ads there.

---

There are probably several ways to make that quantum Boltzmann machine supersymmetric. But it will take a while to see if that can be done while implementing your idea that x = input, theta = learned representations.

---

Yes Porter, it is perhaps unavoidable that there are probably many ways to attain that degree of supersymmetry, as Supermathematics is quite broad.

Unfortunately, my knowledge is very limited, as I lack at minimum a Bachelors physics degree, or any training in physics, so the method outlined in the super Hamiltonian paper above, was the easiest entry point I could garner of based on evidence observed thus far.

---

The Hamiltonian of the transverse Ising model shows up in N=1 SQCD as something like a matrix ("V" in the paper) describing how squark combinations mix. But that's ridiculously abstract for this purpose, there has to be a much simpler way.

---

First, thanks a lot for that source Mitchell.

Secondly, researchers have applied the transverse field Hamiltonian here.( though not supersymmetric) 

Deep abstractions are a common consideration in modern machine learning; maybe replicating our brains in silico requires quite complex equations that generate deep abstractions.

Footnote:

As Max Tegmark expressed in a youtube video here, physicists have long neglected to define the observer in much of the equations. (The observer being the intelligent agent)

Perhaps consciousness may be defined in terms of very complex equations, from disciplines, like Physics; as an example, degrees of general structures such as manifolds central to physics and mathematics, are now quite prevalent in the study of Deep Learning.

---

There is a shortcut in the footnote. Deep learning is made by robots with algorithms of robots even if they hope to mime a human. Consciousness is a concept far from learning and it asks other questions.