Zobrazeno 1 - 10
of 899
pro vyhledávání: '"An, Sanghyeok"'
Autor:
Kim, Hyeonah, Kim, Minsu, Yun, Taeyoung, Choi, Sanghyeok, Bengio, Emmanuel, Hernández-García, Alex, Park, Jinkyoo
Designing biological sequences with desired properties is a significant challenge due to the combinatorially vast search space and the high cost of evaluating each candidate sequence. To address these challenges, reinforcement learning (RL) methods,
Externí odkaz:
http://arxiv.org/abs/2410.04461
Autor:
Kim, Minsu, Choi, Sanghyeok, Yun, Taeyoung, Bengio, Emmanuel, Feng, Leo, Rector-Brooks, Jarrid, Ahn, Sungsoo, Park, Jinkyoo, Malkin, Nikolay, Bengio, Yoshua
Amortized inference is the task of training a parametric model, such as a neural network, to approximate a distribution with a given unnormalized density where exact sampling is intractable. When sampling is implemented as a sequential decision-makin
Externí odkaz:
http://arxiv.org/abs/2410.01432
Autor:
Park, Sanghyeok, Benson, Jared, Corrigan, J., Dodson, J. P., Coppersmith, S. N., Friesen, Mark, Eriksson, M. A.
Latching techniques are widely used to enhance readout of qubits. These methods require precise tuning of multiple tunnel rates, which can be challenging to achieve under realistic experimental conditions, such as when a qubit is coupled to a single
Externí odkaz:
http://arxiv.org/abs/2408.15380
Recent advancements in 3D object detection have benefited from multi-modal information from the multi-view cameras and LiDAR sensors. However, the inherent disparities between the modalities pose substantial challenges. We observe that existing multi
Externí odkaz:
http://arxiv.org/abs/2407.19156
Video Transformers have become the prevalent solution for various video downstream tasks with superior expressive power and flexibility. However, these video transformers suffer from heavy computational costs induced by the massive number of tokens a
Externí odkaz:
http://arxiv.org/abs/2403.13347
Vision Transformer (ViT) has emerged as a prominent backbone for computer vision. For more efficient ViTs, recent works lessen the quadratic cost of the self-attention layer by pruning or fusing the redundant tokens. However, these works faced the sp
Externí odkaz:
http://arxiv.org/abs/2403.10030
We present the Generative Flow Ant Colony Sampler (GFACS), a novel meta-heuristic method that hierarchically combines amortized inference and parallel stochastic search. Our method first leverages Generative Flow Networks (GFlowNets) to amortize a mu
Externí odkaz:
http://arxiv.org/abs/2403.07041
The development of the internet has allowed for the global distribution of content, redefining media communication and property structures through various streaming platforms. Previous studies successfully clarified the factors contributing to trends
Externí odkaz:
http://arxiv.org/abs/2402.19329
Autor:
Kang, Taein, Han, Soyul, Choi, Sunmook, Seo, Jaejin, Chung, Sanghyeok, Lee, Seungeun, Oh, Seungsang, Kwak, Il-Youp
Conventional spoofing detection systems have heavily relied on the use of handcrafted features derived from speech data. However, a notable shift has recently emerged towards the direct utilization of raw speech waveforms, as demonstrated by methods
Externí odkaz:
http://arxiv.org/abs/2402.17127
Autor:
Schug, Daniel, Kovach, Tyler J., Wolfe, M. A., Benson, Jared, Park, Sanghyeok, Dodson, J. P., Corrigan, J., Eriksson, M. A., Zwolak, Justyna P.
The rapid development of quantum dot (QD) devices for quantum computing has necessitated more efficient and automated methods for device characterization and tuning. Many of the measurements acquired during the tuning process come in the form of imag
Externí odkaz:
http://arxiv.org/abs/2402.13699