More Like this

1. Learning Explainable Templated Graphical Model

   Embar, Varun; Srinivasan, Sriram; Getoor, Lise (August 2022, Uncertainty in artificial intelligence)

   Templated graphical models (TGMs) encode model structure using rules that capture recurring relationships between multiple random variables. While the rules in TGMs are interpretable, it is not clear how they can be used to generate explanations for the individual predictions of the model. Further, learning these rules from data comes with high computational costs: it typically requires an expensive combinatorial search over the space of rules and repeated optimization over rule weights. In this work, we propose a new structure learning algorithm, Explainable Structured Model Search (ESMS), that learns a templated graphical model and an explanation framework for its predictions. ESMS uses a novel search procedure to efficiently search the space of models and discover models that trade-off predictive accuracy and explainability. We introduce the notion of relational stability and prove that our proposed explanation framework is stable. Further, our proposed piecewise pseudolikelihood (PPLL) objective does not require re-optimizing the rule weights across models during each iteration of the search. In our empirical evaluation on three realworld datasets, we show that our proposed approach not only discovers models that are explainable, but also significantly outperforms existing state-out-the-art structure learning approaches. 

   more » « less
2. Discovering neural elastoplasticity from kinematic observations

   https://doi.org/10.1073/pnas.2508732122  

   Gavris, Georgios Barkoulis; Sun, WaiChing (September 2025, Proceedings of the National Academy of Sciences)

   Inferring accurate and precise material models necessary for high-fidelity predictions has been a central challenge in constitutive modeling. Both traditional regression methods and modern machine-learning approaches require specialized data labels, which often cannot be sufficiently obtained from experiments. This data demand makes many sophisticated models impractical for real-world problems. Improvements in digital image correlation techniques have enabled accurate measurements of displacement data, providing an alternative kinematics-based approach for model identification. However, for materials undergoing plastic deformation, fracture, and damage, the corresponding inverse problem could be inherently challenging due to the dependence on loading history. We overcome this by formulating an inverse problem to discover interpretable plasticity models parameterized by neural networks (NN) from kinematic observations, leveraging a differentiable simulator with a smooth constitutive update that enables backpropagation for the NN training. The ability to use kinematic observations to infer complex material models may pave the way for a massive generation of material models that can be game-changing for emerging applications such as the design of metamaterials, response surface analyses, and the design of experiments. 

   more » « less
3. Approximating Back-propagation for a Biologically Plausible Local Learning Rule in Spiking Neural Networks

   https://doi.org/10.1145/3354265.3354275  

   Shrestha, Amar; Fang, Haowen; Wu, Qing; Qiu, Qinru (July 2019, International Conference on Neuromorphic Systems)

   Asynchronous event-driven computation and communication using spikes facilitate the realization of spiking neural networks (SNN) to be massively parallel, extremely energy efficient and highly robust on specialized neuromorphic hardware. However, the lack of a unified robust learning algorithm limits the SNN to shallow networks with low accuracies. Artificial neural networks (ANN), however, have the backpropagation algorithm which can utilize gradient descent to train networks which are locally robust universal function approximators. But backpropagation algorithm is neither biologically plausible nor neuromorphic implementation friendly because it requires: 1) separate backward and forward passes, 2) differentiable neurons, 3) high-precision propagated errors, 4) coherent copy of weight matrices at feedforward weights and the backward pass, and 5) non-local weight update. Thus, we propose an approximation of the backpropagation algorithm completely with spiking neurons and extend it to a local weight update rule which resembles a biologically plausible learning rule spike-timing-dependent plasticity (STDP). This will enable error propagation through spiking neurons for a more biologically plausible and neuromorphic implementation friendly backpropagation algorithm for SNNs. We test the proposed algorithm on various traditional and non-traditional benchmarks with competitive results. 

   more » « less
4. Manifold learning for olfactory habituation to strongly fluctuating backgrounds

   https://doi.org/10.1101/2025.05.26.656161  

   Bourassa, François_X P; François, Paul; Reddy, Gautam; Vergassola, Massimo (May 2025, bioRxiv)

   Abstract Animals rely on their sense of smell to survive, but important olfactory cues are mixed with confounding background odors that fluctuate due to atmospheric turbulence. It is unclear how the olfactory system habituates to such stochastic backgrounds to detect behaviorally important odors. Here, we explicitly consider the high-dimensional nature of odor coding, the natural statistics of odor fluctuations and the architecture of the early olfactory pathway. We show that their combination favors a manifold learning mechanism for olfactory habituation over alternatives based on predictive filtering. Manifold learning is implemented in our model by a biologically plausible network of inhibitory interneurons in the early olfactory pathway. We demonstrate that plasticity rules based on IBCM or online PCA are effective at implementing this mechanism in turbulent conditions and outperform previous models relying on mean background subtraction. Interneurons with an IBCM plasticity rule acquire selectivity to independently varying odors. This manifold learning mechanism offers a path towards distinguishing plasticity rules in experiments and could be leveraged by other biological circuits facing fluctuating environments. 

   more » « less
5. Self-Assembly of a Biologically Plausible Learning Circuit

   Liao, Qianli; Liu, Ziyin; Gan, Yulu; Cheung, Brian; Harnett, Mark; Poggio, Tomaso (December 2024, Center for Brains, Minds and Machines (CBMM))

   Over the last four decades, the amazing success of deep learning has been driven by the use of Stochastic Gradient Descent (SGD) as the main optimization technique. The default implementation for the computation of the gradient for SGD is backpropagation, which, with its variations, is used to this day in almost all computer implementations. From the perspective of neuroscientists, however, the consensus is that backpropagation is unlikely to be used by the brain. Though several alternatives have been discussed, none is so far supported by experimental evidence. Here we propose a circuit for updating the weights in a network that is biologically plausible, works as well as backpropagation, and leads to verifiable predictions about the anatomy and the physiology of a characteristic motif of four plastic synapses between ascending and descending cortical streams. A key prediction of our proposal is a surprising property of self-assembly of the basic circuit, emerging from initial random connectivity and heterosynaptic plasticity rules. 

   more » « less