Searcharxiv⌕ Search

arXiv subjects

Wenhao Li

Publications and source records attributed to Wenhao Li.

At least 217 records · Page 12Linked to original sources
arXiv2023

DiffNAS: Bootstrapping Diffusion Models by Prompting for Better Architectures

Diffusion models have recently exhibited remarkable performance on synthetic data. After a diffusion path is selected, a base model, such as UNet, operates as a denoising autoencoder, primarily predicting noises that need to be eliminated step by step. Consequently, it is crucial to employ a model that aligns with the expected budgets to facilitate superior synthetic performance. In this paper, we meticulously analyze the diffusion model and engineer a base model search approach, denoted "DiffNAS". Specifically, we leverage GPT-4 as a supernet to expedite the search, supplemented with a search memory to enhance the results. Moreover, we employ RFID as a proxy to promptly rank the experimental outcomes produced by GPT-4. We also adopt a rapid-convergence training strategy to boost search efficiency. Rigorous experimentation corroborates that our algorithm can augment the search efficiency by 2 times under GPT-based scenarios, while also attaining a performance of 2.82 with 0.37 improvement in FID on CIFAR10 relative to the benchmark IDDPM algorithm.

cs.AI↗
arXiv2023

Semantically Aligned Task Decomposition in Multi-Agent Reinforcement Learning

The difficulty of appropriately assigning credit is particularly heightened in cooperative MARL with sparse reward, due to the concurrent time and structural scales involved. Automatic subgoal generation (ASG) has recently emerged as a viable MARL approach inspired by utilizing subgoals in intrinsically motivated reinforcement learning. However, end-to-end learning of complex task planning from sparse rewards without prior knowledge, undoubtedly requires massive training samples. Moreover, the diversity-promoting nature of existing ASG methods can lead to the "over-representation" of subgoals, generating numerous spurious subgoals of limited relevance to the actual task reward and thus decreasing the sample efficiency of the algorithm. To address this problem and inspired by the disentangled representation learning, we propose a novel "disentangled" decision-making method, Semantically Aligned task decomposition in MARL (SAMA), that prompts pretrained language models with chain-of-thought that can suggest potential goals, provide suitable goal decomposition and subgoal allocation as well as self-reflection-based replanning. Additionally, SAMA incorporates language-grounded RL to train each agent's subgoal-conditioned policy. SAMA demonstrates considerable advantages in sample efficiency compared to state-of-the-art ASG methods, as evidenced by its performance on two challenging sparse-reward tasks, Overcooked and MiniRTS.

cs.LG↗
arXiv2023

Efficient Planning with Latent Diffusion

Temporal abstraction and efficient planning pose significant challenges in offline reinforcement learning, mainly when dealing with domains that involve temporally extended tasks and delayed sparse rewards. Existing methods typically plan in the raw action space and can be inefficient and inflexible. Latent action spaces offer a more flexible paradigm, capturing only possible actions within the behavior policy support and decoupling the temporal structure between planning and modeling. However, current latent-action-based methods are limited to discrete spaces and require expensive planning. This paper presents a unified framework for continuous latent action space representation learning and planning by leveraging latent, score-based diffusion models. We establish the theoretical equivalence between planning in the latent action space and energy-guided sampling with a pretrained diffusion model and incorporate a novel sequence-level exact sampling method. Our proposed method, $\texttt{LatentDiffuser}$, demonstrates competitive performance on low-dimensional locomotion control tasks and surpasses existing methods in higher-dimensional tasks.

cs.LG↗
arXiv2023

Co-Evolution of Pose and Mesh for 3D Human Body Estimation from Video

Despite significant progress in single image-based 3D human mesh recovery, accurately and smoothly recovering 3D human motion from a video remains challenging. Existing video-based methods generally recover human mesh by estimating the complex pose and shape parameters from coupled image features, whose high complexity and low representation ability often result in inconsistent pose motion and limited shape patterns. To alleviate this issue, we introduce 3D pose as the intermediary and propose a Pose and Mesh Co-Evolution network (PMCE) that decouples this task into two parts: 1) video-based 3D human pose estimation and 2) mesh vertices regression from the estimated 3D pose and temporal image feature. Specifically, we propose a two-stream encoder that estimates mid-frame 3D pose and extracts a temporal image feature from the input image sequence. In addition, we design a co-evolution decoder that performs pose and mesh interactions with the image-guided Adaptive Layer Normalization (AdaLN) to make pose and mesh fit the human body shape. Extensive experiments demonstrate that the proposed PMCE outperforms previous state-of-the-art methods in terms of both per-frame accuracy and temporal consistency on three benchmark datasets: 3DPW, Human3.6M, and MPI-INF-3DHP. Our code is available at https://github.com/kasvii/PMCE.

cs.CV↗
arXiv2023

FastLLVE: Real-Time Low-Light Video Enhancement with Intensity-Aware Lookup Table

Low-Light Video Enhancement (LLVE) has received considerable attention in recent years. One of the critical requirements of LLVE is inter-frame brightness consistency, which is essential for maintaining the temporal coherence of the enhanced video. However, most existing single-image-based methods fail to address this issue, resulting in flickering effect that degrades the overall quality after enhancement. Moreover, 3D Convolution Neural Network (CNN)-based methods, which are designed for video to maintain inter-frame consistency, are computationally expensive, making them impractical for real-time applications. To address these issues, we propose an efficient pipeline named FastLLVE that leverages the Look-Up-Table (LUT) technique to maintain inter-frame brightness consistency effectively. Specifically, we design a learnable Intensity-Aware LUT (IA-LUT) module for adaptive enhancement, which addresses the low-dynamic problem in low-light scenarios. This enables FastLLVE to perform low-latency and low-complexity enhancement operations while maintaining high-quality results. Experimental results on benchmark datasets demonstrate that our method achieves the State-Of-The-Art (SOTA) performance in terms of both image quality and inter-frame brightness consistency. More importantly, our FastLLVE can process 1,080p videos at $\mathit{50+}$ Frames Per Second (FPS), which is $\mathit{2 \times}$ faster than SOTA CNN-based methods in inference time, making it a promising solution for real-time applications. The code is available at https://github.com/Wenhao-Li-777/FastLLVE.

cs.CV↗
arXiv2023

Suboptimal subspace construction for log-determinant approximation

Variance reduction is a crucial idea for Monte Carlo simulation and the stochastic Lanczos quadrature method is a dedicated method to approximate the trace of a matrix function. Inspired by their advantages, we combine these two techniques to approximate the log-determinant of large-scale symmetric positive definite matrices. Key questions to be answered for such a method are how to construct or choose an appropriate projection subspace and derive guaranteed theoretical analysis. This paper applies some probabilistic approaches including the projection-cost-preserving sketch and matrix concentration inequalities to construct a suboptimal subspace. Furthermore, we provide some insights on choosing design parameters in the underlying algorithm by deriving corresponding approximation error and probabilistic error estimations. Numerical experiments demonstrate our method's effectiveness and illustrate the quality of the derived error bounds.

math.NA↗
arXiv2023

F2A2: Flexible Fully-decentralized Approximate Actor-critic for Cooperative Multi-agent Reinforcement Learning

Traditional centralized multi-agent reinforcement learning (MARL) algorithms are sometimes unpractical in complicated applications, due to non-interactivity between agents, curse of dimensionality and computation complexity. Hence, several decentralized MARL algorithms are motivated. However, existing decentralized methods only handle the fully cooperative setting where massive information needs to be transmitted in training. The block coordinate gradient descent scheme they used for successive independent actor and critic steps can simplify the calculation, but it causes serious bias. In this paper, we propose a flexible fully decentralized actor-critic MARL framework, which can combine most of actor-critic methods, and handle large-scale general cooperative multi-agent setting. A primal-dual hybrid gradient descent type algorithm framework is designed to learn individual agents separately for decentralization. From the perspective of each agent, policy improvement and value evaluation are jointly optimized, which can stabilize multi-agent policy learning. Furthermore, our framework can achieve scalability and stability for large-scale environment and reduce information transmission, by the parameter sharing mechanism and a novel modeling-other-agents methods based on theory-of-mind and online supervised learning. Sufficient experiments in cooperative Multi-agent Particle Environment and StarCraft II show that our decentralized MARL instantiation algorithms perform competitively against conventional centralized and decentralized methods.

cs.LG↗
arXiv2023

MCPI: Integrating Multimodal Data for Enhanced Prediction of Compound Protein Interactions

The identification of compound-protein interactions (CPI) plays a critical role in drug screening, drug repurposing, and combination therapy studies. The effectiveness of CPI prediction relies heavily on the features extracted from both compounds and target proteins. While various prediction methods employ different feature combinations, both molecular-based and network-based models encounter the common obstacle of incomplete feature representations. Thus, a promising solution to this issue is to fully integrate all relevant CPI features. This study proposed a novel model named MCPI, which is designed to improve the prediction performance of CPI by integrating multiple sources of information, including the PPI network, CCI network, and structural features of CPI. The results of the study indicate that the MCPI model outperformed other existing methods for predicting CPI on public datasets. Furthermore, the study has practical implications for drug development, as the model was applied to search for potential inhibitors among FDA-approved drugs in response to the SARS-CoV-2 pandemic. The prediction results were then validated through the literature, suggesting that the MCPI model could be a useful tool for identifying potential drug candidates. Overall, this study has the potential to advance our understanding of CPI and guide drug development efforts.

q-bio.BM↗
arXiv2023

Temporally-Extended Prompts Optimization for SAM in Interactive Medical Image Segmentation

The Segmentation Anything Model (SAM) has recently emerged as a foundation model for addressing image segmentation. Owing to the intrinsic complexity of medical images and the high annotation cost, the medical image segmentation (MIS) community has been encouraged to investigate SAM's zero-shot capabilities to facilitate automatic annotation. Inspired by the extraordinary accomplishments of interactive medical image segmentation (IMIS) paradigm, this paper focuses on assessing the potential of SAM's zero-shot capabilities within the IMIS paradigm to amplify its benefits in the MIS domain. Regrettably, we observe that SAM's vulnerability to prompt forms (e.g., points, bounding boxes) becomes notably pronounced in IMIS. This leads us to develop a framework that adaptively offers suitable prompt forms for human experts. We refer to the framework above as temporally-extended prompts optimization (TEPO) and model it as a Markov decision process, solvable through reinforcement learning. Numerical experiments on the standardized benchmark BraTS2020 demonstrate that the learned TEPO agent can further enhance SAM's zero-shot capability in the MIS context.

cs.CV↗
arXiv2023

Negotiated Reasoning: On Provably Addressing Relative Over-Generalization

Over-generalization is a thorny issue in cognitive science, where people may become overly cautious due to past experiences. Agents in multi-agent reinforcement learning (MARL) also have been found to suffer relative over-generalization (RO) as people do and stuck to sub-optimal cooperation. Recent methods have shown that assigning reasoning ability to agents can mitigate RO algorithmically and empirically, but there has been a lack of theoretical understanding of RO, let alone designing provably RO-free methods. This paper first proves that RO can be avoided when the MARL method satisfies a consistent reasoning requirement under certain conditions. Then we introduce a novel reasoning framework, called negotiated reasoning, that first builds the connection between reasoning and RO with theoretical justifications. After that, we propose an instantiated algorithm, Stein variational negotiated reasoning (SVNR), which uses Stein variational gradient descent to derive a negotiation policy that provably avoids RO in MARL under maximum entropy policy iteration. The method is further parameterized with neural networks for amortized learning, making computation efficient. Numerical experiments on many RO-challenged environments demonstrate the superiority and efficiency of SVNR compared to state-of-the-art methods in addressing RO.

cs.AI↗
arXiv2023

Real higher-order Weyl photonic crystal

Higher-order Weyl semimetals are a family of recently predicted topological phases simultaneously showcasing unconventional properties derived from Weyl points, such as chiral anomaly, and multidimensional topological phenomena originating from higher-order topology. The higher-order Weyl semimetal phases, with their higher-order topology arising from quantized dipole or quadrupole bulk polarizations, have been demonstrated in phononics and circuits. Here, we experimentally discover a class of higher-order Weyl semimetal phase in a three-dimensional photonic crystal (PhC), exhibiting the concurrence of the surface and hinge Fermi arcs from the nonzero Chern number and the nontrivial generalized real Chern number, respectively, coined a real higher-order Weyl PhC. Notably, the projected two-dimensional subsystem with kz = 0 is a real Chern insulator, belonging to the Stiefel-Whitney class with real Bloch wavefunctions, which is distinguished fundamentally from the Chern class with complex Bloch wavefunctions. Our work offers an ideal photonic platform for exploring potential applications and material properties associated with the higher-order Weyl points and the Stiefel-Whitney class of topological phases.

cond-mat.mes-hall↗
arXiv2023

Machine Learning for Mie-Tronics

Electromagnetic multipole expansion theory underpins nanoscale light-matter interactions, particularly within subwavelength meta-atoms, paving the way for diverse and captivating optical phenomena. While conventionally brute force optimization methods, relying on the iterative exploration of various geometries and materials, are employed to obtain the desired multipolar moments, these approaches are computationally demanding and less effective for intricate designs. In this study, we unveil the potential of machine learning for designing dielectric meta-atoms with desired multipolar moments up to the octupole terms. Specifically, we develop forward prediction models to unravel the intricate relationship between the scattering response and the topological attributes of individual meta-atoms, and an inverse design model to reconstruct scatterers with the targeted multipolar moments. Utilizing a tandem network trained to tailor dielectric meta-atoms for generating intended multipolar moments across a broad spectral range, we further demonstrate the generation of uniquely shaped meta-atoms for exciting exclusive higher order magnetic response and establishing super-scattering regime of light-matter interaction. We also illustrate the accurate prediction of electric field distributions within the given scatterer. Our versatile methodology can be readily applied to existing datasets and seamlessly integrated with various network architectures and problem domains, making it a valuable tool for the design of different platforms at nanoscale.

physics.optics↗
arXiv2023

Efficient Cross-Lingual Transfer for Chinese Stable Diffusion with Images as Pivots

Diffusion models have made impressive progress in text-to-image synthesis. However, training such large-scale models (e.g. Stable Diffusion), from scratch requires high computational costs and massive high-quality text-image pairs, which becomes unaffordable in other languages. To handle this challenge, we propose IAP, a simple but effective method to transfer English Stable Diffusion into Chinese. IAP optimizes only a separate Chinese text encoder with all other parameters fixed to align Chinese semantics space to the English one in CLIP. To achieve this, we innovatively treat images as pivots and minimize the distance of attentive features produced from cross-attention between images and each language respectively. In this way, IAP establishes connections of Chinese, English and visual semantics in CLIP's embedding space efficiently, advancing the quality of the generated image with direct Chinese prompts. Experimental results show that our method outperforms several strong Chinese diffusion models with only 5%~10% training data.

cs.CV↗
arXiv2023

Post-Starburst Properties of Post-Merger Galaxies

Post-starburst galaxies (PSBs) are transition galaxies showing evidence of recent rapid star formation quenching. To understand the role of galaxy mergers in triggering quenching, we investigate the incidence of PSBs and resolved PSB properties in post-merger galaxies using both SDSS single-fiber spectra and MaNGA resolved IFU spectra. We find post-mergers have a PSB excess of 10 - 20 times that relative to their control galaxies using single-fiber PSB diagnostics. A similar excess of ~ 19 times is also found in the fraction of central (C)PSBs and ring-like (R)PSBs in post-mergers using the resolved PSB diagnostic. However, 60% of the CPSBs + RPSBs in both post-mergers and control galaxies are missed by the single-fiber data. By visually inspecting the resolved PSB distribution, we find that the fraction of outside-in quenching is 7 times higher than inside-out quenching in PSBs in post-mergers while PSBs in control galaxies do not show large differences in these quenching directions. In addition, we find a marginal deficit of HI gas in PSBs relative to non-PSBs in post-mergers using the MaNGA-HI data. The excesses of PSBs in post-mergers suggest that mergers play an important role in triggering quenching. Resolved IFU spectra are important to recover the PSBs missed by single-fiber spectra. The excess of outside-in quenching relative to inside-out quenching in post-mergers suggests that AGN are not the dominant quenching mechanism in these galaxies, but that processes from the disk (gas inflows/consumption and stellar feedback) play a more important role.

astro-ph.GA↗
arXiv2023

Deep Learning for Solving and Estimating Dynamic Macro-Finance Models

We develop a methodology that utilizes deep learning to simultaneously solve and estimate canonical continuous-time general equilibrium models in financial economics. We illustrate our method in two examples: (1) industrial dynamics of firms and (2) macroeconomic models with financial frictions. Through these applications, we illustrate the advantages of our method: generality, simultaneous solution and estimation, leveraging the state-of-art machine-learning techniques, and handling large state space. The method is versatile and can be applied to a vast variety of problems.

q-fin.CP↗
arXiv2023

Interweaved Graph and Attention Network for 3D Human Pose Estimation

Despite substantial progress in 3D human pose estimation from a single-view image, prior works rarely explore global and local correlations, leading to insufficient learning of human skeleton representations. To address this issue, we propose a novel Interweaved Graph and Attention Network (IGANet) that allows bidirectional communications between graph convolutional networks (GCNs) and attentions. Specifically, we introduce an IGA module, where attentions are provided with local information from GCNs and GCNs are injected with global information from attentions. Additionally, we design a simple yet effective U-shaped multi-layer perceptron (uMLP), which can capture multi-granularity information for body joints. Extensive experiments on two popular benchmark datasets (i.e. Human3.6M and MPI-INF-3DHP) are conducted to evaluate our proposed method.The results show that IGANet achieves state-of-the-art performance on both datasets. Code is available at https://github.com/xiu-cs/IGANet.

cs.CV↗
arXiv2023

GATOR: Graph-Aware Transformer with Motion-Disentangled Regression for Human Mesh Recovery from a 2D Pose

3D human mesh recovery from a 2D pose plays an important role in various applications. However, it is hard for existing methods to simultaneously capture the multiple relations during the evolution from skeleton to mesh, including joint-joint, joint-vertex and vertex-vertex relations, which often leads to implausible results. To address this issue, we propose a novel solution, called GATOR, that contains an encoder of Graph-Aware Transformer (GAT) and a decoder with Motion-Disentangled Regression (MDR) to explore these multiple relations. Specifically, GAT combines a GCN and a graph-aware self-attention in parallel to capture physical and hidden joint-joint relations. Furthermore, MDR models joint-vertex and vertex-vertex interactions to explore joint and vertex relations. Based on the clustering characteristics of vertex offset fields, MDR regresses the vertices by composing the predicted base motions. Extensive experiments show that GATOR achieves state-of-the-art performance on two challenging benchmarks.

cs.CV↗
arXiv2023

Diverse Policy Optimization for Structured Action Space

Enhancing the diversity of policies is beneficial for robustness, exploration, and transfer in reinforcement learning (RL). In this paper, we aim to seek diverse policies in an under-explored setting, namely RL tasks with structured action spaces with the two properties of composability and local dependencies. The complex action structure, non-uniform reward landscape, and subtle hyperparameter tuning due to the properties of structured actions prevent existing approaches from scaling well. We propose a simple and effective RL method, Diverse Policy Optimization (DPO), to model the policies in structured action space as the energy-based models (EBM) by following the probabilistic RL framework. A recently proposed novel and powerful generative model, GFlowNet, is introduced as the efficient, diverse EBM-based policy sampler. DPO follows a joint optimization framework: the outer layer uses the diverse policies sampled by the GFlowNet to update the EBM-based policies, which supports the GFlowNet training in the inner layer. Experiments on ATSC and Battle benchmarks demonstrate that DPO can efficiently discover surprisingly diverse policies in challenging scenarios and substantially outperform existing state-of-the-art methods.

cs.LG↗