Optimization & Theory

Descending through a Crowded Valley - Benchmarking Deep Learning Optimizers ICML 2021 Determinantal point processes based on orthogonal polynomials for sampling minibatches in SGD NeurIPS 2021 Deterministic Mini-batch Sequencing for Training Deep Neural Networks AAAI 2021 Differentiable Fluids with Solid Coupling for Learning and Control AAAI 2021 Differentiable Spike: Rethinking Gradient-Descent for Training Spiking Neural Networks NeurIPS 2021 Differentiable Spline Approximations NeurIPS 2021 Differential Privacy Dynamics of Langevin Diffusion and Noisy Gradient Descent NeurIPS 2021 DI-Fusion: Online Implicit 3D Reconstruction With Deep Priors CVPR 2021 Direction Matters: On the Implicit Bias of Stochastic Gradient Descent with Moderate Learning Rate ICLR 2021 DISCO: Dynamic and Invariant Sensitive Channel Obfuscation for Deep Neural Networks CVPR 2021 Discretized Integrated Gradients for Explaining Language Models EMNLP 2021 Disentangling Representations of Text by Masking Transformers EMNLP 2021 Dissecting Generation Modes for Abstractive Summarization Models via Ablation and Attribution ACL 2021 Dissecting User-Perceived Latency of On-Device E2E Speech Recognition Interspeech 2021 Distilling Optimal Neural Networks: Rapid Search in Diverse Spaces ICCV 2021 Distribution Adaptive INT8 Quantization for Training CNNs AAAI 2021 Distribution-Aware Adaptive Multi-Bit Quantization CVPR 2021 Diverse Branch Block: Building a Convolution as an Inception-Like Unit CVPR 2021 Diversifying Sample Generation for Accurate Data-Free Quantization CVPR 2021 DNN-based Topology Optimisation: Spatial Invariance and Neural Tangent Kernel NeurIPS 2021 Does Preprocessing Help Training Over-parameterized Neural Networks? NeurIPS 2021 Domain-Independent Dominance of Adaptive Methods CVPR 2021 DOTS: Decoupling Operation and Topology in Differentiable Architecture Search CVPR 2021 Double-Win Quant: Aggressively Winning Robustness of Quantized Deep Neural Networks via Random Precision Training and Inference ICML 2021 DPFPS: Dynamic and Progressive Filter Pruning for Compressing Convolutional Neural Networks from Scratch AAAI 2021