Zobrazeno 1 - 10
of 40
pro vyhledávání: '"Bai, Qinxun"'
Particle-based Bayesian deep learning often requires a similarity metric to compare two networks. However, naive similarity metrics lack permutation invariance and are inappropriate for comparing networks. Centered Kernel Alignment (CKA) on feature k
Externí odkaz:
http://arxiv.org/abs/2411.00259
The improvement of economic policymaking presents an opportunity for broad societal benefit, a notion that has inspired research towards AI-driven policymaking tools. AI policymaking holds the potential to surpass human performance through the abilit
Externí odkaz:
http://arxiv.org/abs/2410.08345
Behavior constrained policy optimization has been demonstrated to be a successful paradigm for tackling Offline Reinforcement Learning. By exploiting historical transitions, a policy is trained to maximize a learned value function while constrained b
Externí odkaz:
http://arxiv.org/abs/2211.15956
While natural gradients have been widely studied from both theoretical and empirical perspectives, we argue that some fundamental theoretical issues regarding the existence of gradients in infinite dimensional function spaces remain underexplored. We
Externí odkaz:
http://arxiv.org/abs/2202.06232
Improving the sample efficiency of reinforcement learning algorithms requires effective exploration. Following the principle of $\textit{optimism in the face of uncertainty}$ (OFU), we train a separate exploration policy to maximize the approximate u
Externí odkaz:
http://arxiv.org/abs/2110.12081
Recently, particle-based variational inference (ParVI) methods have gained interest because they can avoid arbitrary parametric assumptions that are common in variational inference. However, many ParVI approaches do not allow arbitrary sampling from
Externí odkaz:
http://arxiv.org/abs/2103.01291
We propose a novel Siamese Natural Language Tracker (SNLT), which brings the advancements in visual tracking to the tracking by natural language (NL) descriptions task. The proposed SNLT is applicable to a wide range of Siamese trackers, providing a
Externí odkaz:
http://arxiv.org/abs/1912.02048
Efficient exploration remains a challenging problem in reinforcement learning, especially for those tasks where rewards from environments are sparse. A commonly used approach for exploring such environments is to introduce some "intrinsic" reward. In
Externí odkaz:
http://arxiv.org/abs/1911.08017
In recent years, deep-learning-based visual object trackers have been studied thoroughly, but handling occlusions and/or rapid motion of the target remains challenging. In this work, we argue that conditioning on the natural language (NL) description
Externí odkaz:
http://arxiv.org/abs/1907.11751
Conventional unsupervised domain adaptation (UDA) assumes that training data are sampled from a single domain. This neglects the more practical scenario where training data are collected from multiple sources, requiring multi-source domain adaptation
Externí odkaz:
http://arxiv.org/abs/1812.01754