Determinism and Indeterminism as Model Artefacts: Toward a Model-Invariant Ontology of Physics

summary

The gist

The gist: Deterministic–stochastic dualities are available in principle, and arise in a non-contrived way in many scientifically important models.

In short

The paper argues that determinism and indeterminism are not fundamental features of reality but rather artifacts of how we choose to model physics. It proposes a 'Model-Invariant Ontology' criterion: only structural features stable across all empirically equivalent formulations qualify as real. This suggests that the choice between deterministic or stochastic descriptions is a modeling convention, not an ontological truth.

Key concepts

Model-Invariant Ontology Criterion: This criterion dictates that only structural features remain stable when a model is reformulated in an empirically equivalent way. Features like conservation laws are considered real because they persist across different mathematical representations, shifting focus from specific math to persistent relational structures.
Representational Duality and Model-Equivalence: It shows that a deterministic system can be recast as stochastic, and vice versa, while keeping the observable outcomes the same. This duality arises because fine-grained details are lost when coarse-graining for observation. The form of the model (deterministic vs. stochastic) is thus a choice in representation.
Gauge Freedom and Invariant Structure: When state spaces are simplified to match empirical resolution, multiple descriptions—deterministic or stochastic—can yield the same observable dynamics. What is truly invariant across these choices is the underlying modal and relational structure of the system, which should be the focus of ontology.

Terminology used across episodes

Reinforcement Learning (RL) — explained in 24 episodes
Retrieval-Augmented Generation (RAG) — explained in 16 episodes
Interpretability — explained in 15 episodes
Sample Complexity — explained in 15 episodes
Federated Learning — explained in 14 episodes
Generalization — explained in 13 episodes

This episode discusses

The paper

Determinism and Indeterminism as Model Artefacts: Toward a Model-Invariant Ontology of Physics · Read on arXiv

Transcript

Introduction to the show: ident: Quantum Radio. Generated commentary on the latest quantum physics and condensed matter papers.

Kai: Today's paper: "Determinism and Indeterminism as Model Artefacts".

Mira: The gist: Deterministic–stochastic dualities are available in principle, and arise in a non-contrived way in many scientifically important models.

Kai: First, who's behind it and why it matters.

Paper summary: Kai: So, wrapping up this look at "Determinism and Indeterminism as Model Artefacts: Toward a Model-Invariant Ontology of Physics," the core idea is that both determinism and indeterminism are just artifacts of our modeling practice rather than deep features of nature <ref:2512.22540#pg1>.

Mira: Right, and they’re really pushing for this model-invariance criterion where you only count the structural stuff that remains the same no matter how you reformulate the math, like conservation laws or symmetries <ref:2512.22540#pg3>.

Lev: So it’s saying that whether a system is described as deterministic or stochastic just depends on which representation you pick for your specific setup, right?

Kai: That’s the gist of it. Like in quantum mechanics, coarse-graining leads to stochastic rules even if the underlying rules are unitary evolution <ref:2512.22540#pg3>.

Mira: It really shifts the goal of ontology from picking a single theory to finding those robust structural features that survive all those different ways of writing things down <ref:2512.22540#pg3>.

Lev: And for someone working on error correction, this means we focus on what structure is conserved across those transitions, not whether the specific transition itself is deterministic or probabilistic <ref:2512.22540#pg3>.

Kai: It suggests that the distinction between determinism and indeterminism isn't a metaphysical split in the universe itself, but more about how our modeling practice encodes that underlying structure <ref:2512.22540#pg1>.

Mira: It means the real work for physics is figuring out what stays invariant across all those reformulations because that’s where they find what actually matters ontologically <ref:2512.22540#pg3>.

Conclusion: Kai: So, we’ve been digging into this paper about determinism and indeterminism—"Determinism and Indeterminism as Model Artefacts: Toward a Model-Invariant Ontology of Physics"—and the big idea is that both are just artifacts of how we choose to model things instead of deep features in nature.

Mira: Exactly. They propose this model-invariant ontology criterion where you only count structural stuff that stays the same no matter how you reformulate the math, like conservation laws or symmetries.

Lev: So it’s saying whether a system is described as deterministic or stochastic just depends on which representation you pick for your specific setup, right?

Kai: Right. Like in quantum mechanics, coarse-graining leads to stochastic rules even if the underlying rules are unitary evolution. It reframes the measurement problem as a choice of representational framework.

Mira: It really shifts the goal of ontology from picking a single theory to finding those robust structural features that survive all these different ways of writing things down.

Lev: And for someone working on error correction, this means we focus on what structure is conserved across those transitions, not whether the specific transition is deterministic or probabilistic.

Mira: It means the real work for physics is figuring out what stays invariant across all those reformulations. That’s where they find what actually matters ontologically.

Kai: What this does for us is it suggests that instead of getting stuck arguing over which representation is "correct," we should be looking for the mathematical skeleton that all those representations share.

Mira: It means if you can show a transition structure survives coarse-graining, then that structure holds some kind of reality, even if we can't pinpoint the exact microscopic rule governing it.

Lev: For running experiments, this is helpful because it tells us what features we absolutely need to test to make sure our results are independent of how we choose to describe the system.

Kai: It sounds like the paper is saying that when you look at physics, you shouldn't be looking for a single truth about what happens at every point, but rather the stable relations between those points.

Mira: That’s right. It reframes determinism and indeterminism as just different languages we use to describe the same underlying physical reality.

More episodes

← Home