Attention variants, state-space and recurrent cells, normalization and token-mixing schemes — each tested against the standard block it replaces.
Replace Cox-de Boor evaluation of each cubic B-spline edge activation with its fixed truncated-power expansion. Normalize each scalar edge input to a bounded knot coordinate, evaluate the five shifted cubic positive-part terms in parallel, and contract them with the learned spline coefficients inside one fused kernel.
Replace selected residual, recurrent, or state-space blocks by modules whose input-output Jacobians satisfy an IODP inequality throughout a prescribed activation domain. The constraint controls incremental amplification between two trajectories without requiring either trajectory to remain near one fixed equilibrium, so it should improve robustness to changing contexts and prevent exploding long-horizon sensitivities.
Replace independent architecture generation with a diffusion mutation kernel that starts from a known valid neural architecture, re-noises it for only a fraction of the diffusion horizon, and denoises it conditionally toward a new architecture. The resulting candidates should remain closer to the parent and retain validity at low mutation strength, while larger re-noising fractions should produce greater novelty and access to distinct architectural basins.
Construct a residual network from independently attachable modules, but permit only a number of modules whose aggregate feedback gain lies inside a delay-dependent admissible interval. Estimate deployed end-to-end latency and each module's local Jacobian gain, then reject or bypass additional modules when the predicted delayed-loop stability boundary is crossed. This turns variable-width or depth scaling into a falsifiable control problem rather than an empirical choice.
Build a shallow neural model whose input at every discretization level is embedded into a common Hilbert space with uniformly bounded norm, and constrain every neuron parameter in the corresponding dual norm. The statistical complexity then depends on the Hilbert norm bound rather than the number of retained coordinates, allowing one model design to operate across increasingly fine measurements.
Replace black-box differentiation through an embedded LP decision with an analytic Jacobian computed from the LP’s active basis. A neural policy emits LP coefficients or right-hand sides; the LP returns the decision, while the backward pass uses the basis inverse and dual sensitivity, avoiding solver unrolling and finite-difference noise.
Replace hard clipping or post-hoc asymmetric saturation with a dynamic output state that remains inside a prescribed asymmetric interval. A neural network emits a command uc, while the realized output u evolves through the APIR vector field, producing bounded actions, temporal smoothing, and gradients that remain available in the interior.
Replace a single recurrent state update with fast feature relaxation, activity evolution, and a slow adaptive state that modulates the activity vector field. Tune the activity subsystem near a controllable saddle-node so that it retains a useful transient regime for a predictable number of steps, enabling delayed switching and long-horizon memory without requiring a large hidden state.
Replace a standard ReLU surrogate with an input convex neural network whose hidden-to-hidden weights are constrained to be nonnegative. The network remains piecewise linear and expressive, but its convexity allows downstream minimization to use continuous ReLU epigraph constraints instead of binary activation variables, potentially eliminating the integrality bottleneck of neural optimization.
Compute an inner approximation of the states from which a neural controller can keep the plant inside a prescribed safe domain indefinitely, then use the resulting regulation map as a safety shield around the network. At each state, the network proposes an action, but the shield projects or replaces it with an action certified to remain in the invariant set.
Represent each neural module as a Hamiltonian storage system and connect modules through a state-dependent skew or Dirac interconnection instead of arbitrary residual additions. The coupling may change with the hidden state, but its internal power contribution cancels exactly, so total stored energy is controlled only by external inputs and explicitly added dissipation.
Replace a square dense projection in a Transformer or MLP with a trainable recursive butterfly matrix. The layer preserves multiscale channel interactions while constraining every complementary row-column block to rank at most k, reducing parameters and enabling recursive structured matrix-vector products. Unlike an arbitrary sparse layer, the construction has an explicit recursive factorization and a quasi-optimal approximation guarantee among matrices with the same butterfly rank.
Replace a dense neural-network weight tensor with a graph tensor network whose physical modes and internal edge ranks are specified by a sparse rank-adjacency matrix. Unlike tensor-train or hierarchical Tucker layers, the graph can contain selected cycles and skip connections between tensor modes, allowing the factorization topology to match correlations in the weight tensor. Fit the layer with GTN-SVD at a prescribed tolerance and compare accuracy, parameter count, and tensor-contraction…
Replace a single neural representation of a rapidly oscillatory PDE solution by a macroscopic network plus an explicitly oscillatory corrector network. Feed the network both the slow coordinate $x$ and fast coordinate $y=x/\varepsilon$, and train the resulting composite field in a variational energy objective. This targets the paper's scale-robust approximation bound rather than forcing the optimizer and finite sample set to discover oscillations of wavelength $\varepsilon$.
Replace direct learning of a highly cancelling signed observable with a quotient-space model over symmetry orbits of inputs. Predict a physically constrained baseline for each family and use an LSTM or set/graph encoder only for the residual many-body correlation, then aggregate family predictions with known signed weights instead of forming a noisy sample-level ratio.
Partition a neural network into independently trained or independently monitored modules and constrain their cross-module interaction gain using a compositional contraction certificate. This enables stable deep modular MLPs, graph blocks, or recurrent modules without estimating the full network Jacobian, while providing an explicit coupling threshold for when the architecture loses contraction.
Replace a deep feed-forward block by the fixed point z=phi(Wz+Vx+b), with the recurrent weight W constrained so that the fixed point is unique for every input. The same condition makes forward fixed-point iteration stable and makes implicit differentiation well-conditioned, allowing depth-independent memory usage while providing a measurable spectral failure boundary.
Compress each hidden layer by retaining directions that are simultaneously reachable from the observed input distribution and observable at the network output. Unlike PCA or SVD, the retained subspace is weighted by downstream task sensitivity, so high-variance but output-irrelevant directions can be removed while low-variance predictive directions are preserved.
Build a residual module whose state explicitly contains both a persistent context representation and an accumulator. Each residual branch computes one learned correction and adds it to the accumulator, instead of forcing every layer to represent the complete output from scratch. This provides a concrete solver-like architecture for high-dimensional regression and iterative latent prediction.
Construct a continuous-depth or recurrent residual block with a prescribed polynomial Lyapunov decay near its equilibrium. The architecture combines a fixed radial stabilizer with a learned component that is constrained to have zero radial projection, allowing slow algebraic transients and long memory while preventing asymptotic hidden-state growth.
Replace the usual isotropic Gaussian random Fourier features with a frequency distribution matched to the expected spectral regularity of the target function. For coordinate fields, operator-learning maps, or PDE solution surrogates, this should place more features where the target Fourier energy lies and improve approximation at the same feature count. Stabilize the resulting feature matrix with whitening or ridge regression because spectral accuracy can create severe ill-conditioning.
Constrain a student policy to transform its action in the same way that the input state is transformed, while constraining its value estimate to remain unchanged. During distillation, augment every teacher-student pair with several symmetry-transformed copies and penalize disagreement after transforming the student action back to the original frame.
Initialize each row of a neural weight matrix as a stationary correlated Gaussian process instead of using independent entries, but constrain its correlation tail to remain on the finite-fourth-moment side of the transition. This creates controllable structured spectra while avoiding the heavy-edge regime predicted for correlations slower than \(t^{-1/2}\).
Represent a large linear map acting on a Cartesian 3D grid and multiple physical channels as a TT-matrix, while retaining separate TT blocks for channel couplings that have different semantics. Apply the layer by sequential contractions with TT cores rather than materializing a dense matrix or a full 3D convolution kernel. Rank truncation provides an explicit accuracy-versus-memory knob and can be applied after optimizer updates.