Search Results for author:
Found 22 papers, 4 papers with code
On the Convergence of Continual Learning with Adaptive Methods
One of the objectives of continual learning is to prevent catastrophic forgetting in learning multiple tasks sequentially, and the existing solutions have been driven by the conceptualization of the plasticity-stability dilemma.
SPQR: Controlling Q-ensemble Independence with Spiked Random Model for Reinforcement Learning
Alleviating overestimation bias is a critical challenge for deep reinforcement learning to achieve successful performance on more complex tasks or offline datasets containing out-of-distribution data.
Toward Reinforcement Learning-based Rectilinear Macro Placement Under Human Constraints
Macro placement is a critical phase in chip design, which becomes more intricate when involving general rectilinear macros and layout areas.
Pitfall of Optimism: Distributional Reinforcement Learning by Randomizing Risk Criterion
Distributional reinforcement learning algorithms have attempted to utilize estimated uncertainty for exploration, such as optimism in the face of uncertainty.
Distributional Reinforcement Learning
reinforcement-learning
Domain-Aware Fine-Tuning: Enhancing Neural Network Adaptability
By leveraging batch normalization layers and integrating linear probing and fine-tuning, our DAFT significantly mitigates feature distortion and achieves improved model performance on both in-distribution and out-of-distribution datasets.
Learning to Learn Unlearned Feature for Brain Tumor Segmentation
We propose a fine-tuning algorithm for brain tumor segmentation that needs only a few data samples and helps networks not to forget the original tasks.
Bridging the Gap between Model Explanations in Partially Annotated Multi-label Classification
Based on these findings, we propose to boost the attribution scores of the model trained with partial labels to make its explanation resemble that of the model trained with full labels.
Large Loss Matters in Weakly Supervised Multi-Label Classification
In this work, we first regard unobserved labels as negative labels, casting the WSML task into noisy multi-label classification.
Variational Perturbations for Visual Feature Attribution
Explaining a complex black-box system in a post-hoc manner is important to understand its predictions.
Beyond Examples: Constructing Explanation Space for Explaining Prototypes
As deep learning has been successfully deployed in diverse applications, there is ever increasing need for explaining its decision.
Distributional Perturbation for Efficient Exploration in Distributional Reinforcement Learning
Distributional reinforcement learning aims to learn distribution of return under stochastic environments.
Distribution-Driven Disjoint Prediction Intervals for Deep Learning
To address the issue, we propose a novel method that generates a union of disjoint PIs.
On the Convergence of Nonconvex Continual Learning with Adaptive Learning Rate
The memory based continual learning stores a small subset of the data for previous tasks and applies various methods such as quadratic programming and sample selection.
Nonconvex Continual Learning with Episodic Memory
We also show that memory-based approaches have an inherent problem of overfitting to memory, which degrades the performance on previously learned tasks, namely catastrophic forgetting.
Variational saliency maps for explaining model's behavior
Saliency maps have been widely used to explain the behavior of an image classifier.
Information-Theoretic Privacy in Federated Submodel learning
We consider information-theoretic privacy in federated submodel learning, where a global server has multiple submodels.
Information Theory Information Theory
REST: Performance Improvement of a Black Box Model via RL-based Spatial Transformation
Our work aims to improve the robustness by adding a REST module in front of any black boxes and training only the REST module without retraining the original black box model in an end-to-end manner, i. e. we try to convert the real-world data into training distribution which the performance of the black-box model is best suited for.
Adversarial training with perturbation generator networks
However, these adversarial attack methods used in these techniques are fixed, making the model stronger only to attacks used in training, which is widely known as an overfitting problem.
Unsupervised classification into unknown number of classes
We propose a novel unsupervised classification method based on graph Laplacian.
Sampling-based Bayesian Inference with gradient uncertainty
In this paper, we show that predictive uncertainty can be efficiently estimated when we incorporate the concept of gradients uncertainty into posterior sampling.
CTD: Fast, Accurate, and Interpretable Method for Static and Dynamic Tensor Decompositions
Also, CTD-S is made 5 ~ 86x faster, and 7 ~ 12x more memory-efficient than the state-of-the-art method by removing redundancy.
Generative Adversarial Trainer: Defense to Adversarial Perturbations with GAN
The generator network generates an adversarial perturbation that can easily fool the classifier network by using a gradient of each image.
