Learning multiple layers of features from tiny images

representative citing papers

citing papers explorer

Showing 9 of 9 citing papers.

• Dataset Distillation cs.LG · 2018-11-27 · unverdicted · none · ref 98

  Dataset distillation creates a tiny synthetic training set that, when used with a fixed network initialization, produces models whose performance approximates that of models trained on the full original dataset.

• Data Cleansing for Models Trained with SGD stat.ML · 2019-06-20 · unverdicted · none · ref 10

  A method infers influential instances for data cleansing in SGD-trained models by retracing training steps with intermediate models, shown to improve performance on MNIST and CIFAR10.

• Adaptive Federated Optimization cs.LG · 2020-02-29 · unverdicted · none · ref 223

  Proposes federated adaptive optimizers (FedAdagrad, FedAdam, FedYogi) with convergence analysis for non-convex objectives under data heterogeneity and reports empirical gains over FedAvg.

• Learnable Parameter Similarity cs.LG · 2019-07-27 · unverdicted · none · ref 7

  LPS uses a second-order neural network to learn an end-to-end metric for second-order parameter similarity and introduces the ModelSet500 benchmark with 500 trained models.

• Graph-based Knowledge Distillation by Multi-head Attention Network cs.LG · 2019-07-04 · unverdicted · none · ref 14

  Multi-head attention constructs a graph of dataset relations from the teacher embedding procedure and transfers it to the student via multi-task learning, yielding 7.05% higher CIFAR-100 accuracy than the student alone and 2.46% above prior SOTA.

• EPNAS: Efficient Progressive Neural Architecture Search cs.LG · 2019-07-07 · unverdicted · none · ref 23

  EPNAS uses a progressive search policy with REINFORCE performance prediction to search neural architectures in parallel, supporting multiple resource constraints and outperforming ENAS and PNAS on CIFAR-10 and ImageNet in speed and accuracy.

• Beneficial perturbation network for continual learning cs.LG · 2019-06-22 · unverdicted · none · ref 19

  BPN adds task-specific beneficial perturbations as biases to neural networks to overcome catastrophic forgetting without storing prior data or expanding the network substantially.

• Submodular Batch Selection for Training Deep Neural Networks cs.LG · 2019-06-20 · unverdicted · none · ref 13

  A greedy submodular maximization method for mini-batch selection in DNN training yields better generalization than SGD on standard datasets.

• Weight Normalization based Quantization for Deep Neural Network Compression cs.LG · 2019-07-01 · unverdicted · none · ref 18

  WNQ uses weight normalization to reshape weight distributions and reduce quantization error, outperforming baselines on CIFAR-100 and ImageNet.