A new realistic stochastic interpretation of Quantum Mechanics

[Fra]

Sure but can you or Barandes provide a simple example of INMS processes? What does it look like?

Again a simple example of how that memory effect emerges would be great.

Depends on what you think is simple, but I think finance agents is simple and graspable.

Lets say we consider the dynamics of the market value of a stock. Then a simple model is that the market value is somehow determined by the collective expectations of the agents. And each agent can have different levels of sophisticated predictions, but an agent with memory will somehow have a memory depth and use part of history, in it's evaluation of the expectation. This means that the probability of a given value tomorrow, does not depend only on what it is today, but alsowhat is was in the past. But realistically an agent does not remember or know ALL of history, it consider or retains only part of it. This can be called "memory depth", but one can also imagine agents where the memory depth relates to its capacity, and thus every agent as a lossy retention of the past; which influence the expectations of the future.

So the market memory is this collective effect where the agents memory of thet past affects the valuations. This makes the transition from value to another emergent and non-markovian.

Non-divisible is because the emergent dynamics depends critically on the collective behaviour of agents including feedback loops. Arbitrarily dividing it into individual possible strategies and them just average them will miss the feedback between agent; and agents vs the collective.

This naturally has a cooperative element, as any part of the system is dependent on it's immediate surrounding. So even a "selfish agent" cant ignore that it depends on its environment. This is where emergence and selforganisation comes in.

are you proposing some sort of "hard" emergent macroproperty that cannot be explained from the micro theory?

The above is the simple idea. But the critical part to take this to next level is HOW the agents revise their expectations based on the history. This is a more deep water that goes beyond the basic question of getting a handle on what "indivisibility" and non-markovian can mean. If we mix this discussion with the meaning of non-disible and non-markobian i think it will be a mess. The agents memory can a simple model as an explicit actual memory or it can be implicit as in i the agents doe reinfored learning; thus the agent itself may well evolve and change behaviour (learn), this is an "implicit memory". But still makes in non-markovian.

a random references... but finance papers arent asking the ame questions as we do, so the analogies are never perfect...
Non-Markovian Dynamics for Automated Trading

Edit: Forgot to mention the obvious as well, that the "memory" of the agent, is in a sense "subjective hidden varibles" that does not satsfiy bell assumptions. And one can IMO very well consider hte agents actions even to be partly stochastic. But the market dynamics is an interaction of these processes. That si the point.

I think this is "easy" as it requires no "weirdness". the only challenge, is to "translate" agent processing, agent encoding, agent phenomenolgoy to "physics". But this is just a "framework". no mistakes needs to be done to think that any of this implies that particles "think" like human, or process inforamtion like humans. It can be "implicit", and even related to evolutionary learning, and thus "memory effects".

/Fredrik

 

[JC_Silver]

Another noob question, as always, how would QFT look under a INMS interpretation? If particles have definite position/momentum, how would field theories behave? The same? Different? Better? Worse?
I ask because QFT is a powerful tool that assumes particles are a result of disturbances on a field and not classical particles, as far as my education took me.

 

[PAllen]

Another noob question, as always, how would QFT look under a INMS interpretation? If particles have definite position/momentum, how would field theories behave? The same? Different? Better? Worse?
I ask because QFT is a powerful tool that assumes particles are a result of disturbances on a field and not classical particles, as far as my education took me.

Barandes has claimed, but to my knowledge, never published on this, that his approach generalizes to infinite configurations, and that then his dictionary approach maps to QFT.

 

[Lord Jestocost]

I'm not gonna lie, this is what gets me about the non-Markovian process, because the dice rolls isn't a non-Markovian process, it's a regular stochastic process. Since I'm not well versed in stochastic processes of any kind and I can't find good sources on non-Markovian processes online that are not by Barandes (as we used to say in the long past, my Google-fu isn't strong enough), I'm left not understanding exactly why the particle has definite positions between division events, because as far as I understand, the dice roll also doesn't exist between one roll and the next.

Again, sorry for bothering >.<

When modeling a physical system, one should reflect whether the physical model was restricted to a subset of the system’s phase space to analyze its behavior. When only a subset of the system's coordinates was examined, the system's behavior might appear "non-Markovian", thus the notation “apparent non-Markovian” would be more suitable.

 

[jbergman]

Depends on what you think is simple, but I think finance agents is simple and graspable.

Lets say we consider the dynamics of the market value of a stock. Then a simple model is that the market value is somehow determined by the collective expectations of the agents. And each agent can have different levels of sophisticated predictions, but an agent with memory will somehow have a memory depth and use part of history, in it's evaluation of the expectation. This means that the probability of a given value tomorrow, does not depend only on what it is today, but alsowhat is was in the past. But realistically an agent does not remember or know ALL of history, it consider or retains only part of it. This can be called "memory depth", but one can also imagine agents where the memory depth relates to its capacity, and thus every agent as a lossy retention of the past; which influence the expectations of the future.

So the market memory is this collective effect where the agents memory of thet past affects the valuations. This makes the transition from value to another emergent and non-markovian.

This still sounds markovian to me if each agent has a finite memory. Markovian processes aren't necessarily just one step conditionally independent.

Non-divisible is because the emergent dynamics depends critically on the collective behaviour of agents including feedback loops. Arbitrarily dividing it into individual possible strategies and them just average them will miss the feedback between agent; and agents vs the collective.

This seems like the more important point than the non-markovian claim you made above.

.
This naturally has a cooperative element, as any part of the system is dependent on it's immediate surrounding. So even a "selfish agent" cant ignore that it depends on its environment. This is where emergence and selforganisation comes in.

The above is the simple idea. But the critical part to take this to next level is HOW the agents revise their expectations based on the history. This is a more deep water that goes beyond the basic question of getting a handle on what "indivisibility" and non-markovian can mean. If we mix this discussion with the meaning of non-disible and non-markobian i think it will be a mess. The agents memory can a simple model as an explicit actual memory or it can be implicit as in i the agents doe reinfored learning; thus the agent itself may well evolve and change behaviour (learn), this is an "implicit memory". But still makes in non-markovian.

a random references... but finance papers arent asking the ame questions as we do, so the analogies are never perfect...
Non-Markovian Dynamics for Automated Trading

I have to read this paper and reply back. So far, I haven't found much of this discussion that illuminating. I found Barendes own video much more grounded and interesting. One comment that I found stimulating in that video from a colleague of Barendes was that indivisibility does fundamentally imply non-locality. Barendes said that may be true but that we just don't have an intuition about indivisibility.