Constrained Co-Design for Photonic Bayesian Neural Networks

• URL: http://arxiv.org/abs/2608.02229v1
• Date: Mon, 03 Aug 2026 13:47:48 GMT
• Title: Constrained Co-Design for Photonic Bayesian Neural Networks
• Abstract summary: Bayesian neural networks (BNNs) provide a principled framework for uncertainty-aware prediction by replacing deterministic parameters with probability distributions.<n>Photonic computing offers a promising alternative by exploiting intrinsic opticality for fast and parallel sampling.<n>We study which hardware-imposed constraints limit photonic BNN inference, how these constraints can be represented, and which ranges can be tolerated by photonic BNNs beyond small proof-of-concept networks.
• Score: 4.191204698939203
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Classical neural networks frequently produce overconfident predictions on ambiguous or out-of-distribution (OOD) data, a liability that grows with each AI system deployed in safety-critical real-world scenarios. Bayesian neural networks (BNNs) provide a principled framework for uncertainty-aware prediction by replacing deterministic parameters with probability distributions, but repeated sampling increases latency, memory traffic, and energy consumption. Photonic probabilistic computing offers a promising alternative by exploiting intrinsic optical stochasticity for fast and parallel sampling. However, photonic BNNs are not ideal samplers: analog constraints on quantization, programming error, dynamic range, and representable mean and variance restrict the variational families that can be implemented in hardware. In this work, we study which hardware-imposed constraints limit scalable photonic BNN inference, how these constraints can be represented, and which ranges can be tolerated by photonic BNNs beyond small proof-of-concept networks. We formulate photonic BNN inference as constrained stochastic variational inference and perform a systematic ablation study over stochasticity location, stochasticity modality, quantization, programming error, and mean/variance bounds. From these results, we derive concrete co-design guidelines that distinguish hardware constraints that can be compensated by training from those requiring hardware or architecture intervention. We validate these guidelines under coupled, hardware-realistic constraints on Dirty-MNIST, CIFAR-10, and CINIC-10, using Fashion-MNIST and SVHN as OOD benchmarks, showing that hardware-aware training recovers predictive performance and uncertainty quality whenever the required variational family remains representable, whereas violations of representational limits require targeted hardware modifications.

Related papers

• Disentangling Aleatoric and Epistemic Uncertainty in Physics-Informed Neural Networks. Application to Insulation Material Degradation Prognostics [1.1417805445492082]
  Most existing PIN-based prognostics approaches are deterministic or account only for uncertainty, limiting their suitability for risk-aware decision-making.<n>This work introduces a heteroscedastic Bayesian Physics-Informed Neural Network (B-PINN) framework that integrates BNNs with physics-based residual enforcement and prior distributions.<n>Results show that the proposed B-PINN provides improved predictive accuracy and better systematic uncertainty estimates than competing approaches.
  arXiv  Detail & Related papers  (2026-01-07T07:54:09Z)
• Uncertainty-Preserving QBNNs: Multi-Level Quantization of SVI-Based Bayesian Neural Networks for Image Classification [9.160739594332036]
  We introduce a systematic multi-level quantization framework for Variational Inference based BNNs.<n>We demonstrate that BNNs can be quantized down to 4-bit precision while maintaining both classification accuracy and uncertainty disentanglement.
  arXiv  Detail & Related papers  (2025-12-11T12:51:42Z)
• A Conformal Prediction Framework for Uncertainty Quantification in Physics-Informed Neural Networks [0.9613011825024471]
  PINNs have emerged as a powerful framework for solving PDEs.<n>Existing uncertainty quantification approaches for PINNs generally lack rigorous statistical guarantees.<n>We introduce a distribution-free conformal prediction framework for UQ in PINNs.
  arXiv  Detail & Related papers  (2025-09-17T06:12:48Z)
• End-to-End Probabilistic Framework for Learning with Hard Constraints [43.297232073870866]
  We present ProbHardE2E, a probabilistic forecasting framework that incorporates hard operational/physical constraints.<n>Our methodology uses a novel differentiable probabilistic projection layer (DPPL) that can be combined with a wide range of neural network architectures.
  arXiv  Detail & Related papers  (2025-06-08T05:29:50Z)
• Evaluating Single Event Upsets in Deep Neural Networks for Semantic Segmentation: an embedded system perspective [1.474723404975345]
  This paper delves into the robustness assessment in embedded Deep Neural Networks (DNNs)<n>By scrutinizing the layer-by-layer and bit-by-bit sensitivity of various encoder-decoder models to soft errors, this study thoroughly investigates the vulnerability of segmentation DNNs to SEUs.<n>We propose a set of practical lightweight error mitigation techniques with no memory or computational cost suitable for resource-constrained deployments.
  arXiv  Detail & Related papers  (2024-12-04T18:28:38Z)
• Bayesian Entropy Neural Networks for Physics-Aware Prediction [14.705526856205454]
  We introduce BENN, a framework designed to impose constraints on Bayesian Neural Network (BNN) predictions. Benn is capable of constraining not only the predicted values but also their derivatives and variances, ensuring a more robust and reliable model output. Results highlight significant improvements over traditional BNNs and showcase competitive performance relative to contemporary constrained deep learning methods.
  arXiv  Detail & Related papers  (2024-07-01T07:00:44Z)
• Using Uncertainty Quantification to Characterize and Improve Out-of-Domain Learning for PDEs [44.890946409769924]
  Neural Operators (NOs) have emerged as particularly promising quantification. We show that ensembling several NOs can identify high-error regions and provide good uncertainty estimates. We then introduce Operator-ProbConserv, a method that uses these well-calibrated UQ estimates within the ProbConserv framework to update the model.
  arXiv  Detail & Related papers  (2024-03-15T19:21:27Z)
• Achieving Constraints in Neural Networks: A Stochastic Augmented Lagrangian Approach [49.1574468325115]
  Regularizing Deep Neural Networks (DNNs) is essential for improving generalizability and preventing overfitting. We propose a novel approach to DNN regularization by framing the training process as a constrained optimization problem. We employ the Augmented Lagrangian (SAL) method to achieve a more flexible and efficient regularization mechanism.
  arXiv  Detail & Related papers  (2023-10-25T13:55:35Z)
• Random-Set Neural Networks (RS-NN) [4.549947259731147]
  We propose a novel Random-Set Neural Network (RS-NN) approach to classification.<n> RS-NN encodes the 'epistemic' uncertainty induced by training sets that are insufficiently representative.<n>Our approach outperforms state-of-the-art Bayesian and Ensemble methods in terms of accuracy, uncertainty estimation and out-of-distribution (OoD) detection.
  arXiv  Detail & Related papers  (2023-07-11T20:00:35Z)
• On the Forward Invariance of Neural ODEs [92.07281135902922]
  We propose a new method to ensure neural ordinary differential equations (ODEs) satisfy output specifications. Our approach uses a class of control barrier functions to transform output specifications into constraints on the parameters and inputs of the learning system.
  arXiv  Detail & Related papers  (2022-10-10T15:18:28Z)
• Comparative Analysis of Interval Reachability for Robust Implicit and Feedforward Neural Networks [64.23331120621118]
  We use interval reachability analysis to obtain robustness guarantees for implicit neural networks (INNs) INNs are a class of implicit learning models that use implicit equations as layers. We show that our approach performs at least as well as, and generally better than, applying state-of-the-art interval bound propagation methods to INNs.
  arXiv  Detail & Related papers  (2022-04-01T03:31:27Z)
• NUQ: Nonparametric Uncertainty Quantification for Deterministic Neural Networks [151.03112356092575]
  We show the principled way to measure the uncertainty of predictions for a classifier based on Nadaraya-Watson's nonparametric estimate of the conditional label distribution. We demonstrate the strong performance of the method in uncertainty estimation tasks on a variety of real-world image datasets.
  arXiv  Detail & Related papers  (2022-02-07T12:30:45Z)
• Certification of Iterative Predictions in Bayesian Neural Networks [79.15007746660211]
  We compute lower bounds for the probability that trajectories of the BNN model reach a given set of states while avoiding a set of unsafe states. We use the lower bounds in the context of control and reinforcement learning to provide safety certification for given control policies.
  arXiv  Detail & Related papers  (2021-05-21T05:23:57Z)
• Chance-Constrained Control with Lexicographic Deep Reinforcement Learning [77.34726150561087]
  This paper proposes a lexicographic Deep Reinforcement Learning (DeepRL)-based approach to chance-constrained Markov Decision Processes. A lexicographic version of the well-known DeepRL algorithm DQN is also proposed and validated via simulations.
  arXiv  Detail & Related papers  (2020-10-19T13:09:14Z)
• Robustness Analysis of Neural Networks via Efficient Partitioning with Applications in Control Systems [45.35808135708894]
  Neural networks (NNs) are now routinely implemented on systems that must operate in uncertain environments. This paper unifies propagation and partition approaches to provide a family of robustness analysis algorithms. New partitioning techniques are aware of their current bound estimates and desired boundary shape.
  arXiv  Detail & Related papers  (2020-10-01T16:51:36Z)
• Frequentist Uncertainty in Recurrent Neural Networks via Blockwise Influence Functions [121.10450359856242]
  Recurrent neural networks (RNNs) are instrumental in modelling sequential and time-series data. Existing approaches for uncertainty quantification in RNNs are based predominantly on Bayesian methods. We develop a frequentist alternative that: (a) does not interfere with model training or compromise its accuracy, (b) applies to any RNN architecture, and (c) provides theoretical coverage guarantees on the estimated uncertainty intervals.
  arXiv  Detail & Related papers  (2020-06-20T22:45:32Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.