1 code implementation • 3 Oct 2024 • Seungyong Moon, Bumsoo Park, Hyun Oh Song
To this end, we propose guided stream of search (GSoS), which seamlessly incorporates optimal solutions into the self-generation process in a progressive manner, producing high-quality search trajectories.
1 code implementation • 29 Aug 2024 • Jang-Hyun Kim, Claudia Skok Gibbs, Sangdoo Yun, Hyun Oh Song, Kyunghyun Cho
We propose a novel machine learning approach for inferring causal variables of a target variable from observations.
1 code implementation • 21 Jun 2024 • Deokjae Lee, Hyun Oh Song, Kyunghyun Cho
Active learning is increasingly adopted for expensive multi-objective combinatorial optimization problems, but it involves a challenging subset selection problem, optimizing the batch acquisition score that quantifies the goodness of a batch for evaluation.
1 code implementation • 18 Jun 2024 • Jinuk Kim, Marwa El Halabi, Mingi Ji, Hyun Oh Song
Existing depth compression methods remove redundant non-linear activation functions and merge the consecutive convolution layers into a single layer.
1 code implementation • 6 Dec 2023 • Jang-Hyun Kim, Junyoung Yeom, Sangdoo Yun, Hyun Oh Song
This paper presents a context key/value compression method for Transformer language models in online scenarios, where the context continually expands.
1 code implementation • NeurIPS 2023 • Seungyong Moon, Junyoung Yeom, Bumsoo Park, Hyun Oh Song
Discovering achievements with a hierarchical structure in procedurally generated environments presents a significant challenge.
1 code implementation • 27 May 2023 • Deokjae Lee, JunYeong Lee, Jung-Woo Ha, Jin-Hwa Kim, Sang-Woo Lee, Hwaran Lee, Hyun Oh Song
To this end, we propose Bayesian red teaming (BRT), novel query-efficient black-box red teaming methods based on Bayesian optimization, which iteratively identify diverse positive test cases leading to model failures by utilizing the pre-defined user input pool and the past evaluations.
1 code implementation • NeurIPS 2023 • Gaon An, Junhyeok Lee, Xingdong Zuo, Norio Kosaka, Kyung-Min Kim, Hyun Oh Song
We apply our algorithm to offline RL tasks with actual human preference labels and show that our algorithm outperforms or is on par with the existing PbRL methods.
1 code implementation • NeurIPS 2023 • Jang-Hyun Kim, Sangdoo Yun, Hyun Oh Song
To this end, we present scalable and effective algorithms for detecting label errors and outlier data based on the relational graph structure of data.
1 code implementation • 28 Jan 2023 • Jinuk Kim, Yeonwoo Jeong, Deokjae Lee, Hyun Oh Song
We propose a subset selection problem that replaces inefficient activation layers with identity functions and optimally merges consecutive convolution operations into shallow equivalent convolution operations for efficient end-to-end inference latency.
1 code implementation • 18 Oct 2022 • Seungyong Moon, JunYeong Lee, Hyun Oh Song
Our work focuses on training RL agents on multiple visually diverse environments to improve observational generalization performance.
1 code implementation • 17 Jun 2022 • Deokjae Lee, Seungyong Moon, Junhyeok Lee, Hyun Oh Song
We focus on the problem of adversarial attacks against models on discrete sequential data in the black-box setting where the attacker aims to craft adversarial examples with limited query access to the victim model.
2 code implementations • 30 May 2022 • Jang-Hyun Kim, Jinuk Kim, Seong Joon Oh, Sangdoo Yun, Hwanjun Song, JoonHyun Jeong, Jung-Woo Ha, Hyun Oh Song
The great success of machine learning with massive amounts of data comes at a price of huge computation costs and storage for training and tuning.
no code implementations • 24 Feb 2022 • Yeonwoo Jeong, Deokjae Lee, Gaon An, Changyong Son, Hyun Oh Song
We first show the greedy approach of recent channel pruning methods ignores the inherent quadratic coupling between channels in the neighboring layers and cannot safely remove inactive weights during the pruning procedure.
1 code implementation • 10 Dec 2021 • Seungyong Moon, Gaon An, Hyun Oh Song
However, the vulnerability of neural networks against adversarial attacks poses a serious threat to the people affected by these systems.
5 code implementations • NeurIPS 2021 • Gaon An, Seungyong Moon, Jang-Hyun Kim, Hyun Oh Song
However, prior methods typically require accurate estimation of the behavior policy or sampling from OOD data points, which themselves can be a non-trivial problem.
Ranked #1 on Gym halfcheetah-random on D4RL
1 code implementation • ICLR 2021 • Jang-Hyun Kim, Wonho Choo, Hosan Jeong, Hyun Oh Song
While deep neural networks show great performance on fitting to the training distribution, improving the networks' generalization performance to the test distribution and robustness to the sensitivity to input perturbations still remain as a challenge.
no code implementations • 1 Jan 2021 • Yeonwoo Jeong, Deokjae Lee, Gaon An, Changyong Son, Hyun Oh Song
Reducing the heavy computational cost of large convolutional neural networks is crucial when deploying the networks to resource-constrained environments.
no code implementations • 1 Jan 2021 • Seungyong Moon, Gaon An, Hyun Oh Song
Recent advances on adversarial defense mainly focus on improving the classifier’s robustness against adversarially perturbed inputs.
1 code implementation • ICML 2020 • Jang-Hyun Kim, Wonho Choo, Hyun Oh Song
While deep neural networks achieve great performance on fitting the training distribution, the learned networks are prone to overfitting and are susceptible to adversarial attacks.
Ranked #2 on Image Classification on Tiny-ImageNet
1 code implementation • 23 May 2019 • Yeonwoo Jeong, Hyun Oh Song
Furthermore, we propose a method which avoids offloading the entire burden of jointly modeling the continuous and discrete factors to the variational encoder by employing a separate discrete inference procedure.
1 code implementation • 16 May 2019 • Seungyong Moon, Gaon An, Hyun Oh Song
Solving for adversarial examples with projected gradient descent has been demonstrated to be highly effective in fooling the neural network based classifiers.
no code implementations • CVPR 2019 • Yeonwoo Jeong, Yoonsung Kim, Hyun Oh Song
We develop hierarchically quantized efficient embedding representations for similarity-based search and show that this representation provides not only the state of the art performance on the search accuracy but also provides several orders of speed up during inference.
1 code implementation • 2 Oct 2018 • Hyoungseok Kim, Jaekyeom Kim, Yeonwoo Jeong, Sergey Levine, Hyun Oh Song
Reinforcement learning algorithms struggle when the reward signal is very sparse.
no code implementations • 27 Sep 2018 • HyoungSeok Kim, Jaekyeom Kim, Yeonwoo Jeong, Sergey Levine, Hyun Oh Song
Policy optimization struggles when the reward feedback signal is very sparse and essentially becomes a random search algorithm until the agent stumbles upon a rewarding or the goal state.
1 code implementation • ICML 2018 • Yeonwoo Jeong, Hyun Oh Song
To this end, we consider the problem of directly learning a quantizable embedding representation and the sparse binary hash code end-to-end which can be used to construct an efficient hash table not only providing significant search reduction in the number of data but also achieving the state of the art search accuracy outperforming previous state of the art deep metric learning methods.
1 code implementation • 30 Mar 2017 • Alireza Fathi, Zbigniew Wojna, Vivek Rathod, Peng Wang, Hyun Oh Song, Sergio Guadarrama, Kevin P. Murphy
We propose a new method for semantic instance segmentation, by first computing how likely two pixels are to belong to the same object, and then by grouping similar pixels together.
1 code implementation • CVPR 2017 • Hyun Oh Song, Stefanie Jegelka, Vivek Rathod, Kevin Murphy
Learning the representation and the similarity metric in an end-to-end fashion with deep networks have demonstrated outstanding results for clustering and retrieval.
no code implementations • NeurIPS 2016 • Ozan Sener, Hyun Oh Song, Ashutosh Saxena, Silvio Savarese
Supervised learning with large scale labelled datasets and deep layered models has caused a paradigm shift in diverse areas in learning and recognition.
no code implementations • 10 Feb 2016 • Ozan Sener, Hyun Oh Song, Ashutosh Saxena, Silvio Savarese
We incorporate the domain shift and the transductive target inference into our framework by jointly solving for an asymmetric similarity metric and the optimal transductive target label assignment.
3 code implementations • CVPR 2016 • Hyun Oh Song, Yu Xiang, Stefanie Jegelka, Silvio Savarese
Additionally, we collected Online Products dataset: 120k images of 23k classes of online products for metric learning.
no code implementations • NeurIPS 2014 • Hyun Oh Song, Yong Jae Lee, Stefanie Jegelka, Trevor Darrell
The increasing prominence of weakly labeled data nurtures a growing demand for object detection methods that can cope with minimal supervision.
no code implementations • 28 May 2014 • Tim Althoff, Hyun Oh Song, Trevor Darrell
While low-level image features have proven to be effective representations for visual recognition tasks such as object recognition and scene classification, they are inadequate to capture complex semantic meaning required to solve high-level visual tasks such as multimedia event detection and recognition.
no code implementations • 5 Mar 2014 • Hyun Oh Song, Ross Girshick, Stefanie Jegelka, Julien Mairal, Zaid Harchaoui, Trevor Darrell
Learning to localize objects with minimal supervision is an important problem in computer vision, since large fully annotated datasets are extremely costly to obtain.
Ranked #41 on Weakly Supervised Object Detection on PASCAL VOC 2007