no code implementations • 27 Dec 2023 • Xianyi Chen, Fazhan Liu, Dong Jiang, Kai Yan
Recently, some research show that deep neural networks are vulnerable to the adversarial attacks, the well-trainned samples or patches could be used to trick the neural network detector or human visual perception.
1 code implementation • 2 Nov 2023 • Kai Yan, Alexander G. Schwing, Yu-Xiong Wang
To address this problem, we propose Primal Wasserstein DICE (PW-DICE), which minimizes the primal Wasserstein distance between the expert and learner state occupancies with a pessimistic regularizer and leverages a contrastively learned distance as the underlying metric for the Wasserstein distance.
1 code implementation • 6 Oct 2023 • Andy Zhou, Kai Yan, Michal Shlapentokh-Rothman, Haohan Wang, Yu-Xiong Wang
While large language models (LLMs) have demonstrated impressive performance on a range of decision-making tasks, they rely on simple acting processes and fall short of broad deployment as autonomous agents.
Ranked #3 on Code Generation on HumanEval
no code implementations • 6 Jul 2023 • Kai Yan, Fujun Luan, Miloš Hašan, Thibault Groueix, Valentin Deschaintre, Shuang Zhao
A 3D digital scene contains many components: lights, materials and geometries, interacting to reach the desired appearance.
no code implementations • ICCV 2023 • Cheng Sun, Guangyan Cai, Zhengqin Li, Kai Yan, Cheng Zhang, Carl Marshall, Jia-Bin Huang, Shuang Zhao, Zhao Dong
In the last stage, initialized by the neural predictions, we perform PBIR to refine the initial results and obtain the final high-quality reconstruction of object shape, material, and illumination.
Ranked #1 on Depth Prediction on Stanford-ORB
1 code implementation • 18 Oct 2022 • Kai Yan, Alexander G. Schwing, Yu-Xiong Wang
To better benefit from available demonstrations, we develop a method to Combine Explicit and Implicit Priors (CEIP).
no code implementations • NeurIPS 2021 • Kai Yan, Jie Yan, Chuan Luo, Liting Chen, QIngwei Lin, Dongmei Zhang
Prediction+optimization is a common real-world paradigm where we have to predict problem parameters before solving the optimization problem.
1 code implementation • 22 Nov 2021 • Kai Yan, Jie Yan, Chuan Luo, Liting Chen, QIngwei Lin, Dongmei Zhang
Prediction+optimization is a common real-world paradigm where we have to predict problem parameters before solving the optimization problem.
no code implementations • 14 Aug 2021 • Zhenggang Tang, Kai Yan, Liting Sun, Wei Zhan, Changliu Liu
To efficiently simulate with massive amounts of agents in MPS, we propose Scalable Million-Agent DQN (SMADQN).
no code implementations • 7 Oct 2020 • Yun Cao, Yuebin Wang, Junhuan Peng, Liqiang Zhang, Linlin Xu, Kai Yan, Lihua Li
With a small number of labeled samples for training, it can save considerable manpower and material resources, especially when the amount of high spatial resolution remote sensing images (HSR-RSIs) increases considerably.
1 code implementation • 6 Feb 2020 • Christoph Dlapa, Johannes Henn, Kai Yan
Differential equations are a powerful tool for evaluating Feynman integrals.
High Energy Physics - Phenomenology High Energy Physics - Theory Mathematical Physics Mathematical Physics
no code implementations • 17 Jan 2020 • Yunlong Lu, Kai Yan
Deep reinforcement learning (RL) has achieved outstanding results in recent years, which has led a dramatic increase in the number of methods and applications.