The neural radiance field (NeRF) has shown promising results in preserving the fine details of objects and scenes. However, unlike mesh-based representations, it remains an open problem to build dense correspondences across different NeRFs of the same category, which is essential in many downstream tasks. The main difficulties of this problem lie in the implicit nature of NeRF and the lack of ground-truth correspondence annotations. In this paper, we show it is possible to bypass these challenges by leveraging the rich semantics and structural priors encapsulated in a pre-trained NeRF-based GAN. Specifically, we exploit such priors from three aspects, namely 1) a dual deformation field that takes latent codes as global structural indicators, 2) a learning objective that regards generator features as geometric-aware local descriptors, and 3) a source of infinite object-specific NeRF samples. Our experiments demonstrate that such priors lead to 3D dense correspondence that is accurate, smooth, and robust. We also show that established dense correspondence across NeRFs can effectively enable many NeRF-based downstream applications such as texture transfer.

Deep latent variable models have achieved significant empirical successes in model-based reinforcement learning (RL) due to their expressiveness in modeling complex transition dynamics. On the other hand, it remains unclear theoretically and empirically how latent variable models may facilitate learning, planning, and exploration to improve the sample efficiency of RL. In this paper, we provide a representation view of the latent variable models for state-action value functions, which allows both tractable variational learning algorithm and effective implementation of the optimism/pessimism principle in the face of uncertainty for exploration. In particular, we propose a computationally efficient planning algorithm with UCB exploration by incorporating kernel embeddings of latent variable models. Theoretically, we establish the sample complexity of the proposed approach in the online and offline settings. Empirically, we demonstrate superior performance over current state-of-the-art algorithms across various benchmarks.

StyleGAN has achieved great progress in 2D face reconstruction and semantic editing via image inversion and latent editing. While studies over extending 2D StyleGAN to 3D faces have emerged, a corresponding generic 3D GAN inversion framework is still missing, limiting the applications of 3D face reconstruction and semantic editing. In this paper, we study the challenging problem of 3D GAN inversion where a latent code is predicted given a single face image to faithfully recover its 3D shapes and detailed textures. The problem is ill-posed: innumerable compositions of shape and texture could be rendered to the current image. Furthermore, with the limited capacity of a global latent code, 2D inversion methods cannot preserve faithful shape and texture at the same time when applied to 3D models. To solve this problem, we devise an effective self-training scheme to constrain the learning of inversion. The learning is done efficiently without any real-world 2D-3D training pairs but proxy samples generated from a 3D GAN. In addition, apart from a global latent code that captures the coarse shape and texture information, we augment the generation network with a local branch, where pixel-aligned features are added to faithfully reconstruct face details. We further consider a new pipeline to perform 3D view-consistent editing. Extensive experiments show that our method outperforms state-of-the-art inversion methods in both shape and texture reconstruction quality. Code and data will be released.

We present 3DHumanGAN, a 3D-aware generative adversarial network (GAN) that synthesizes images of full-body humans with consistent appearances under different view-angles and body-poses. To tackle the representational and computational challenges in synthesizing the articulated structure of human bodies, we propose a novel generator architecture in which a 2D convolutional backbone is modulated by a 3D pose mapping network. The 3D pose mapping network is formulated as a renderable implicit function conditioned on a posed 3D human mesh. This design has several merits: i) it allows us to harness the power of 2D GANs to generate photo-realistic images; ii) it generates consistent images under varying view-angles and specifiable poses; iii) the model can benefit from the 3D human prior. Our model is adversarially learned from a collection of web images needless of manual annotation.

Score-based modeling through stochastic differential equations (SDEs) has provided a new perspective on diffusion models, and demonstrated superior performance on continuous data. However, the gradient of the log-likelihood function, i.e., the score function, is not properly defined for discrete spaces. This makes it non-trivial to adapt \textcolor{\cdiff}{the score-based modeling} to categorical data. In this paper, we extend diffusion models to discrete variables by introducing a stochastic jump process where the reverse process denoises via a continuous-time Markov chain. This formulation admits an analytical simulation during backward sampling. To learn the reverse process, we extend score matching to general categorical data and show that an unbiased estimator can be obtained via simple matching of the conditional marginal distributions. We demonstrate the effectiveness of the proposed method on a set of synthetic and real-world music and image benchmarks.

In offline reinforcement learning (RL), a learner leverages prior logged data to learn a good policy without interacting with the environment. A major challenge in applying such methods in practice is the lack of both theoretically principled and practical tools for model selection and evaluation. To address this, we study the problem of model selection in offline RL with value function approximation. The learner is given a nested sequence of model classes to minimize squared Bellman error and must select among these to achieve a balance between approximation and estimation error of the classes. We propose the first model selection algorithm for offline RL that achieves minimax rate-optimal oracle inequalities up to logarithmic factors. The algorithm, ModBE, takes as input a collection of candidate model classes and a generic base offline RL algorithm. By successively eliminating model classes using a novel one-sided generalization test, ModBE returns a policy with regret scaling with the complexity of the minimally complete model class. In addition to its theoretical guarantees, it is conceptually simple and computationally efficient, amounting to solving a series of square loss regression problems and then comparing relative square loss between classes. We conclude with several numerical simulations showing it is capable of reliably selecting a good model class.

通过区分真实和合成样品，鉴别器在训练生成对抗网络（GAN）中起着至关重要的作用。尽管实际数据分布保持不变，但由于发电机的发展，合成分布一直变化，从而影响鉴别器的BI分类任务的相应变化。我们认为，对其容量进行即时调整的歧视者可以更好地适应这种时间变化的任务。一项全面的实证研究证实，所提出的培训策略称为Dynamicd，改善了合成性能，而不会产生任何其他计算成本或培训目标。在不同的数据制度下开发了两个容量调整方案，用于培训gan：i）给定足够数量的培训数据，歧视者从逐渐增加的学习能力中受益，ii）ii）当培训数据受到限制时，逐渐减少层宽度的宽度减轻。歧视者的过度问题。在一系列数据集上进行的2D和3D感知图像合成任务的实验证实了我们的动力学的普遍性以及对基准的实质性改进。此外，Dynamicd与其他歧视器改进方法（包括数据增强，正规化器和预训练）具有协同作用，并且在将学习gans合并时会带来连续的性能增长。

表示学习通常通过管理维度的诅咒在加强学习中起关键作用。代表性的算法类别利用了随机过渡动力学的光谱分解，以构建在理想化环境中具有强大理论特性的表示。但是，当前的光谱方法的适用性有限，因为它们是用于仅国家的聚合并源自策略依赖性过渡内核的，而无需考虑勘探问题。为了解决这些问题，我们提出了一种替代光谱方法，光谱分解表示（SPEDER），该方法从动力学中提取了国家行动抽象而不诱导虚假依赖数据收集策略，同时还可以平衡探索访问权分析交易 - 在学习过程中关闭。理论分析确定了在线和离线设置中所提出的算法的样本效率。此外，一项实验研究表明，在几个基准测试中，比当前的最新算法表现出色。

图像和语言建模对于视觉前训练（VLP）至关重要，该培训旨在从大规模配对的图像文本数据中学习多模式表示。但是，我们观察到，大多数现有的VLP方法着重于建模图像和文本特征之间的相互作用，同时忽略图像和文本之间的信息差异，从而遭受焦点偏见。为了解决这个问题，我们提出了一个视觉语言掩盖自动编码器框架（VLMAE）。VLMAE采用视觉生成学习，促进该模型获得细粒度和公正的特征。与以前的作品不同，Vlmae注意图像中几乎所有关键的补丁，提供了更全面的理解。广泛的实验表明，VLMAE在各种视觉语言下游任务中取得更好的性能，包括视觉问答，即使有20％的预训练速度，图像文本检索和视觉接地也是如此。

在混合完成的多任务，多域和多模式数据上进行预训练仍然是视力感知预训练的开放挑战。在本文中，我们提出了GPPF，这是一个普遍的感知预训练框架，预先培训任务级的动态网络，该网络是由在标签的多任务和多域数据集上的各层知识“乐高”组成的。通过检查人类在复杂环境中学习的先天能力，我们识别并将三个关键要素转移到深网上：（1）同时暴露于每个批次中的各种交叉任务和跨域信息。 （2）由知识共享驱动的单独的乐高单元中的分区知识存储。 （3）用于训练和下游任务的乐高单元子集的稀疏激活。值得注意的是，由于其在输入形状，损失功能，输出格式，数据分布等方面的差异，不同视觉任务的联合培训是不平凡的。因此，我们创新地开发了插件的多任务培训算法，该培训算法是支持单个迭代多个任务（SIMT）同时培训。 Simt用大型多任务多任务数据集为预训练的基础奠定了基础，并且被证明对于我们的GPPF实验中的稳定培训至关重要。令人兴奋的是，详尽的实验表明，我们的GPPF-R50型号在GPPF-15M中的8个预训练预培训任务的强大基线上取得了显着改善，并在22个下游任务中收获了一系列SOTA，并具有相似的计算预算。我们还验证了GPPF对SOTA视觉变压器的概括能力，并具有一致的改进。这些可靠的实验结果充分证明了我们新颖的GPPF框架提供的有效的知识学习，存储，共享和转移。