---
title: 'ST-SSM: Spatiotemporal State Space Neural Operator'
url: https://www.emergentmind.com/topics/spatiotemporal-state-space-neural-operator-st-ssm
type: topic
---

# ST-SSM: Spatiotemporal State Space Neural Operator

A Spatiotemporal State Space Neural Operator (ST-SSM) is a framework designed for learning operators that map spatiotemporal function-valued inputs to outputs, with primary application to modeling and surrogate learning of time-dependent partial differential equations (PDEs), partial integro-differential equations (PIDEs), and high-dimensional dynamical fields under control or external forcing. The core principle is the unification of state-space models (SSMs) for temporal evolution with operator-learning techniques for spatial structure, yielding architectures that integrate memory, stability, high expressivity, and parameter efficiency for complex spatiotemporal systems.

## 1. Mathematical Foundation and Operator Formulation

ST-SSMs are grounded in a synthesis of classical control-theoretic state-space models and neural operator paradigms. The general form encompasses both continuous-time and discrete-time updates:

Continuous-time SSM:
\[
\frac{d}{dt} v(t) = A\,v(t) + B\,u(t), \qquad y(t) = C\,v(t) + D\,u(t)
\]
where $v(t)$ is the hidden state, $u(t)$ the system input (often a spatial field), $y(t)$ the output, and $A, B, C, D$ are learned or input-dependent system matrices/functions [2507.23428], [2410.02113].

In spatially extended systems, the input $u$ is a field $u(x,t)$ with $x$ denoting spatial coordinates and $t$ time. A Galerkin-based projection is often used to map $u(\cdot,t)$ to a finite-dimensional latent state:
\[
u(x,t) \approx \sum_{i=1}^r a_i(t) \phi_i(x)
\]
where $\{\phi_i\}$ are neural-network-parameterized spatial basis functions [2603.17867].

In state-space neural operator frameworks, the operator learning goal is posed as direct approximation of the solution (or flow map) for PDEs/PIDEs:
\[
u(\cdot, t) = \mathcal{S}[u_0, f(\cdot, \cdot)](x, t)
\]
for systems such as
\[
\frac{\partial u(x,t)}{\partial t} = N[u(\cdot,t)](x) + f(x,t)
\]
with $N$ a nonlinear operator (e.g., nonlocal kernel or neural field term) and $f$ the control/exogenous input [2603.17867].

## 2. Architectural Components and Model Instantiations

ST-SSM architectures can be largely decomposed into the following classes:

- **Galerkin-projected neural ODE/PIDE solvers**: These learn both the spatial basis (via a neural network) and the reduced-order temporal ODE system, often bypassing explicit integration by directly parameterizing the flow map as a function of initial state and projected inputs [2603.17867].

- **Structured spatiotemporal scans**: In models such as Mamba Neural Operator and canonical ST-SSMs, structured SSMs are applied along temporal and spatial axes, with factorized spatial SSMs handling distinct dimensions (via bidirectional sweeps or spatial SSM blocks) and temporal SSMs implementing memory and non-Markovian dynamics [2507.23428], [2410.02113].

- **Selective/Adaptive SSM blocks**: ST-Mamba, STG-Mamba, and FR-Mamba instantiate SSM cores whose transition, input, and output kernels are input-dependent and may be gated or fused with graph neural network (GNN) or Fourier-operator branches, often exploiting per-step or per-feature dynamic adaptation for efficiency and stability [2404.13257], [2403.12418], [2505.16083].

- **Biological or neuro-dynamical extensions**: The Parallelized Hierarchical Connectome (PHC) framework extends SSMs to spiking, hierarchical, and biophysically-constrained domains by separating neuron-intrinsic diagonal recurrences from spatial synaptic interactions and supporting advanced neural constraints such as Dale’s Law, short-term plasticity, and reward-modulated STDP [2604.01295].

## 3. Training Methodologies and Flow Learning

Many ST-SSM frameworks employ two-phase or composite losses:

- **Basis function learning (POD-supervision)**: For architectures using modal projections, spatial bases are learned to approximate empirical principal components or dominant singular vectors of snapshot matrices, using orthonormality and basis-approximation losses [2603.17867].

- **Operator (flow map) learning**: The projected state evolution (often an ODE or RNN) is not integrated numerically, but mapped directly via neural nets (e.g., RNNs, LSTMs, Mamba modules) that learn the flow map from initial coefficients and input projections, with loss computed as mean squared error between predicted and true spatiotemporal fields [2603.17867], [2507.23428], [2410.02113].

- **Input-dependent parameterization**: Input-adaptive gates and SSM kernels, as in ST-Mamba and STG-Mamba, allow dynamic adaptation of step sizes, transition matrices, and output kernels, often employing softplus nonlinearities or learned Kalman-fusion for multi-scale data [2404.13257], [2403.12418].

- **Parallel scan and complexity advantages**: Modern diagonal/parallel SSM blocks exploit scan algorithms (with O(log T) parallel depth) for temporal and/or spatial recurrence, yielding efficient implementations with linear complexity in sequence or field size, avoiding quadratic complexity of attention and other token-mixing mechanisms [2310.19694], [2604.01295].

## 4. Theoretical Properties: Universality, Stability, and Expressivity

A distinguishing feature of the ST-SSM paradigm is rigorous theoretical support:

- **Universality**: Convolutional neural operators with SSM-based kernels are universal approximators of continuous spatiotemporal operators, provided their full field-of-view criterion is satisfied (i.e., spatial kernels can propagate information globally) [2507.23428]. The Stone–Weierstrass theorem guarantees that such architectures can approximate any continuous operator on compact sets.

- **Stability and Causality**: By leveraging the state-space structure, ST-SSMs inherit control-theoretic properties. Eigenvalue spectra of A matrices may be regularized (spectral radius or contractivity) to ensure stability, and causality is structurally enforced via recurrence and the form of temporal convolution kernels (e.g., lower-triangular or convolutional in time) [2410.02113].

- **Interpretation as Neural Operators**: Learned Green’s functions in SSMs provide a principled and expressive mechanism for instantiating convolutional integral operators in both space and time (continuous or discrete), generalizing over classical Fourier neural operators, and enabling adaptive locality or globality [2507.23428], [2410.02113], [2603.17867].

## 5. Empirical Benchmarks, Comparison, and Efficiency

ST-SSM models demonstrate strong empirical performance and computational efficiency across domains. Key results include:

- **PDE Benchmarks**: On tasks such as 1D Kuramoto–Sivashinsky, viscous Burgers, and 2D Navier–Stokes, ST-SSM outperforms FNOs, S4-FFNO, and CNN baselines in normalized $\ell_2$ relative error while using $10$–$300\times$ fewer parameters. For instance, at $128$ spatial resolution, ST-SSM achieves $0.0086$ error (vs $0.0317$ S4-FFNO) on KS ($\nu=0.075$) and $0.0345$ vs $0.0477$ (FNO2D) on 2D Navier–Stokes [2507.23428].

- **Efficiency and Scaling**: ST-SSM enables linear scaling in sequence or field size, substantial speedups over Transformer-based attention (e.g., 61% faster than SOTA Transformer on traffic flow, matching or exceeding accuracy), and constant-time autoregressive prediction desirable for physical forecasting and real-time control [2404.13257], [2310.19694].

- **Transfer Learning and Adaptability**: In control-theoretic operator frameworks such as RHYME-XT, partial fine-tuning on novel data (e.g., new kernels in neural field equations) yields fast convergence and accuracy with minimal retraining, supporting generalization across data regimes [2603.17867].

## 6. Extensions, Ablations, and Practical Considerations

- **Spatial Factorization and Bidirectionality**: Empirical ablations highlight the necessity of bidirectional spatial sweeps for universal expressivity and sharp error optima. Unidirectional (causal) spatial SSMs are non-universal and plateau in error [2507.23428].

- **Role of Temporal Memory**: Under conditions of partial observability or unknown system parameters, temporal memory in the SSM acts as a sufficient statistic, inferring missing context and maintaining performance robustness [2507.23428].

- **Adaptive and Hybrid Blocks**: Several instantiations combine SSMs with Fourier (FNO), convolutional operators, or graph neural modules (e.g., STG-Mamba’s fusion with Kalman-filtered GNN features) to cover heterogeneous or multi-modal spatiotemporal data [2403.12418], [2505.16083].

- **Biological Constraints**: PHC-style ST-SSMs show that complex bio-physical priors—including leaky integrate-and-fire, short-term plasticity, and reward modulation—can be imposed without loss of efficiency, suggesting a principled path to parameter-efficient and biologically plausible sequence models [2604.01295].

## 7. Summary Table: Model Attributes and Experimental Regimes

| Architecture       | Spatiotemporal Mix  | Complexity      | Empirical Regimes                  |
|--------------------|--------------------|-----------------|------------------------------------|
| RHYME-XT           | Learned Galerkin flow | Two-stage, projection + flow | Neural field PIDEs                 |
| MNO/ST-SSM [Mamba] | Diagonal/parallel SSM scan | O(L), linear in field size | Standard PDE benchmarks, data-scarcity |
| ST-Mamba           | Selective state/adaptivity | O(TN), linear in nodes | Traffic flow forecasting           |
| STG-Mamba          | Kalman-GNN adaptivity + SSM | O(T), linear | Spatial-temporal graph, system embedding |
| ConvSSM/ConvS5     | Convolutional SSM  | O(L), subquadratic          | Long-range video, AR environments  |
| PHC/PHCSSM         | Diagonal SSM + multi-transmission | O(log T), biologically plausible | Physiological time series    |

Each instantiation validates the general principle that spatiotemporal state-space operators offer an efficient, robust, and theoretically principled architecture for learning complex solution maps of modern time-dependent systems [2603.17867], [2410.02113], [2507.23428], [2310.19694], [2404.13257], [2403.12418], [2604.01295], [2505.16083].

Source: https://www.emergentmind.com/topics/spatiotemporal-state-space-neural-operator-st-ssm