Tag
Proposes a Center of Gravity guided weight correction method for fault-tolerant deep neural networks, achieving significant fault tolerance improvements on LSTM and CNN models without retraining.
Proposes HDE-Net, a manifold-constrained deep neural network that uses hyperbolic space to better model rule-based structures in tabular data, achieving state-of-the-art performance on the TALENT-tiny-core benchmark while maintaining efficiency.
AdaStop is a cost-aware early stopping framework for DNN test selection that optimally stops labeling when the marginal fault discovery rate falls below a threshold, achieving 65-84% fault discovery using only 9-31% of the labeling budget.
This paper proposes a controllability–observability framework for compressing deep neural networks by reducing hidden-state redundancy, demonstrating significant compression with minimal accuracy loss on MNIST and CIFAR-10.
This paper introduces Evidential Adversarial Training (EV-AT), a method that improves the robustness-uncertainty trade-off in classifiers by combining an evidence-based loss with robust evidence alignment, achieving state-of-the-art results on selective classification benchmarks.
Stanford CS229 online course announcement covering building LLMs, deep neural networks, TensorFlow, Keras, OpenCV, and NLP with spaCy.
This paper applies explainable AI techniques (SHAP, SSHAP) to deep neural network models to analyze drivers of electricity prices across 39 European bidding zones, finding that solar power and gas prices are key drivers despite solar's lower generation share.
The paper proposes that grokking in deep neural networks arises from noise-driven escape from metastable phases in first-order L2 phase transitions, demonstrating that delayed generalization follows Arrhenius scaling and reproduces canonical grokking curves.
This paper proposes a novel structured neuron pruning framework for deep neural networks using multi-armed bandit algorithms, demonstrating effectiveness on various tasks.
This paper establishes a theoretical framework showing that smooth activations in deep neural networks can mitigate the curse of dimensionality in uniform convergence, providing non-asymptotic guarantees and outperforming ReLU networks in worst-case reliability.
This paper presents a Marchenko-Pastur random matrix approach to pruning deep neural networks, offering theoretical guarantees and achieving strong accuracy retention with minimal fine-tuning on ImageNet for ViT and CNN architectures.
The paper proposes an attack-agnostic robustness metric based on the spectral norm of the Fisher Information Matrix, providing theoretical bounds and scalable evaluation methods for deep neural networks.
A mathematically principled framework, Variational Inference Evidential Deep Learning (VI-EDL), is proposed to address limitations in conventional Evidential Deep Learning by reformulating it through variational inference, deriving an Evidence Lower Bound, establishing a generalization bound, and achieving state-of-the-art performance on visual and medical datasets.
This paper introduces CAFD, a learning-based approach for DNN fault detection that integrates model-based, distance-based, and a novel concept-based feature called Concept Failure Ratio (CFR) derived from Vision-Language Models. CAFD consistently outperforms state-of-the-art baselines in fault detection rate across multiple datasets and budgets.
This paper presents a comprehensive analysis of the Neural Tangent Generalization Attack (NTGA) for data protection, including a taxonomy of related attacks, and discusses future research directions.