arXiv subjects

Hao Liu

Publications and source records attributed to Hao Liu.

At least 433 records · Page 24Linked to original sources

arXiv2023

Guide Your Agent with Adaptive Multimodal Rewards

Developing an agent capable of adapting to unseen environments remains a difficult challenge in imitation learning. This work presents Adaptive Return-conditioned Policy (ARP), an efficient framework designed to enhance the agent's generalization ability using natural language task descriptions and pre-trained multimodal encoders. Our key idea is to calculate a similarity between visual observations and natural language instructions in the pre-trained multimodal embedding space (such as CLIP) and use it as a reward signal. We then train a return-conditioned policy using expert demonstrations labeled with multimodal rewards. Because the multimodal rewards provide adaptive signals at each timestep, our ARP effectively mitigates the goal misgeneralization. This results in superior generalization performances even when faced with unseen text instructions, compared to existing text-conditioned policies. To improve the quality of rewards, we also introduce a fine-tuning method for pre-trained multimodal encoders, further enhancing the performance. Video demonstrations and source code are available on the project website: \url{https://sites.google.com/view/2023arp}.

cs.LG↗

arXiv2023

UUKG: Unified Urban Knowledge Graph Dataset for Urban Spatiotemporal Prediction

Accurate Urban SpatioTemporal Prediction (USTP) is of great importance to the development and operation of the smart city. As an emerging building block, multi-sourced urban data are usually integrated as urban knowledge graphs (UrbanKGs) to provide critical knowledge for urban spatiotemporal prediction models. However, existing UrbanKGs are often tailored for specific downstream prediction tasks and are not publicly available, which limits the potential advancement. This paper presents UUKG, the unified urban knowledge graph dataset for knowledge-enhanced urban spatiotemporal predictions. Specifically, we first construct UrbanKGs consisting of millions of triplets for two metropolises by connecting heterogeneous urban entities such as administrative boroughs, POIs, and road segments. Moreover, we conduct qualitative and quantitative analysis on constructed UrbanKGs and uncover diverse high-order structural patterns, such as hierarchies and cycles, that can be leveraged to benefit downstream USTP tasks. To validate and facilitate the use of UrbanKGs, we implement and evaluate 15 KG embedding methods on the KG completion task and integrate the learned KG embeddings into 9 spatiotemporal models for five different USTP tasks. The extensive experimental results not only provide benchmarks of knowledge-enhanced USTP models under different task settings but also highlight the potential of state-of-the-art high-order structure-aware UrbanKG embedding methods. We hope the proposed UUKG fosters research on urban knowledge graphs and broad smart city applications. The dataset and source code are available at https://github.com/usail-hkust/UUKG/.

cs.AI↗

arXiv2023

Chain of Hindsight Aligns Language Models with Feedback

Learning from human preferences is important for language models to match human needs and to align with human and social values. Prior works have achieved remarkable successes by learning from human feedback to understand and follow instructions. Nonetheless, these methods are either founded on hand-picked model generations that are favored by human annotators, rendering them inefficient in terms of data utilization and challenging to apply in general, or they depend on reinforcement learning, which often suffers from imperfect reward functions and relies on extremely challenging optimizations. In this work, we propose a novel technique, Chain of Hindsight, that is easy to optimize and can learn from any form of feedback, regardless of its polarity. Our idea is inspired by how humans learn from extensive feedback presented in the form of languages. We convert all types of feedback into sequences of sentences, which are then used to fine-tune the model, allowing us to take advantage of the language comprehension capabilities of language models. We condition the model on a sequence of model generations paired with feedback. By doing so, the model is trained to generate outputs based on feedback, while learning to identify and correct negative attributes or errors. Applying our method to large language models, we observed that Chain of Hindsight significantly surpasses previous methods in aligning language models with human preferences. We report significant improvements on summarization and dialogue benchmarks, with our approach markedly preferred in human evaluations.

cs.LG↗

arXiv2023

Connections between Operator-splitting Methods and Deep Neural Networks with Applications in Image Segmentation

Deep neural network is a powerful tool for many tasks. Understanding why it is so successful and providing a mathematical explanation is an important problem and has been one popular research direction in past years. In the literature of mathematical analysis of deep neural networks, a lot of works is dedicated to establishing representation theories. How to make connections between deep neural networks and mathematical algorithms is still under development. In this paper, we give an algorithmic explanation for deep neural networks, especially in their connections with operator splitting. We show that with certain splitting strategies, operator-splitting methods have the same structure as networks. Utilizing this connection and the Potts model for image segmentation, two networks inspired by operator-splitting methods are proposed. The two networks are essentially two operator-splitting algorithms solving the Potts model. Numerical experiments are presented to demonstrate the effectiveness of the proposed networks.

cs.CV↗

arXiv2023

Multi-omics Sampling-based Graph Transformer for Synthetic Lethality Prediction

Synthetic lethality (SL) prediction is used to identify if the co-mutation of two genes results in cell death. The prevalent strategy is to abstract SL prediction as an edge classification task on gene nodes within SL data and achieve it through graph neural networks (GNNs). However, GNNs suffer from limitations in their message passing mechanisms, including over-smoothing and over-squashing issues. Moreover, harnessing the information of non-SL gene relationships within large-scale multi-omics data to facilitate SL prediction poses a non-trivial challenge. To tackle these issues, we propose a new multi-omics sampling-based graph transformer for SL prediction (MSGT-SL). Concretely, we introduce a shallow multi-view GNN to acquire local structural patterns from both SL and multi-omics data. Further, we input gene features that encode multi-view information into the standard self-attention to capture long-range dependencies. Notably, starting with batch genes from SL data, we adopt parallel random walk sampling across multiple omics gene graphs encompassing them. Such sampling effectively and modestly incorporates genes from omics in a structure-aware manner before using self-attention. We showcase the effectiveness of MSGT-SL on real-world SL tasks, demonstrating the empirical benefits gained from the graph transformer and multi-omics data.

cs.LG↗

arXiv2023

Inverse design of a pyrochlore lattice of DNA origami through model-driven experiments

Sophisticated statistical mechanics approaches and human intuition have demonstrated the possibility to self-assemble complex lattices or finite size constructs, but have mostly only been successful in silico. The proposed strategies quite often fail in experiment due to unpredicted traps associated to kinetic slowing down (gelation, glass transition), as well as to competing ordered structures. An additional challenge that theoretical predictions face is the difficulty to encode the desired inter-particle interaction potential with the currently available library of nano- and micron-sized particles. To overcome these issues, we conjugate here SAT-assembly -- a patchy-particle interaction design algorithm based on constrained optimization solvers -- with coarse-grained simulations of DNA nanotechnology to experimentally realize trap-free self-assembly pathways. As a proof of concept we investigate the assembly of the pyrochlore (also known as tetrastack) lattice, a highly coveted 3D crystal lattice due to its promise in construction of optical metamaterials. We confirm the successful assembly with two different patchy DNA origami designs via SAXS as well as SEM visualization of the silica-coated lattice. Our approach offers a versatile modeling pipeline that starts from patchy particles designed in silico and ends with wireframe DNA origami that self-assemble into the desired structure.

cond-mat.soft↗

arXiv2023

Experimental observation of the significant difference between surface and bulk Kondo processes in Kondo lattice YbCu$_2$Si$_2$

Synchrotron-based angle-resolved photoemission spectroscopy was employed to investigate the temperature evolution of the Yb 4f spectral for surface and bulk in the Kondo lattice YbCu$_2$Si$_2$.Our study quantitatively distinguishes between the surface and bulk hybridization processes, revealing that the onset temperatures for both surface and bulk hybridization processes are significantly higher than the Kondo temperature. Additionally, we found that the effective surface Kondo temperature is much lower than that of the bulk. These findings offer valuable insights into the understanding of heavy fermion physics.

cond-mat.str-el↗

arXiv2023

Machine Learning for Urban Air Quality Analytics: A Survey

The increasing air pollution poses an urgent global concern with far-reaching consequences, such as premature mortality and reduced crop yield, which significantly impact various aspects of our daily lives. Accurate and timely analysis of air pollution is crucial for understanding its underlying mechanisms and implementing necessary precautions to mitigate potential socio-economic losses. Traditional analytical methodologies, such as atmospheric modeling, heavily rely on domain expertise and often make simplified assumptions that may not be applicable to complex air pollution problems. In contrast, Machine Learning (ML) models are able to capture the intrinsic physical and chemical rules by automatically learning from a large amount of historical observational data, showing great promise in various air quality analytical tasks. In this article, we present a comprehensive survey of ML-based air quality analytics, following a roadmap spanning from data acquisition to pre-processing, and encompassing various analytical tasks such as pollution pattern mining, air quality inference, and forecasting. Moreover, we offer a systematic categorization and summary of existing methodologies and applications, while also providing a list of publicly available air quality datasets to ease the research in this direction. Finally, we identify several promising future research directions. This survey can serve as a valuable resource for professionals seeking suitable solutions for their specific challenges and advancing their research at the cutting edge.

cs.LG↗

arXiv2023

Cloth2Body: Generating 3D Human Body Mesh from 2D Clothing

In this paper, we define and study a new Cloth2Body problem which has a goal of generating 3D human body meshes from a 2D clothing image. Unlike the existing human mesh recovery problem, Cloth2Body needs to address new and emerging challenges raised by the partial observation of the input and the high diversity of the output. Indeed, there are three specific challenges. First, how to locate and pose human bodies into the clothes. Second, how to effectively estimate body shapes out of various clothing types. Finally, how to generate diverse and plausible results from a 2D clothing image. To this end, we propose an end-to-end framework that can accurately estimate 3D body mesh parameterized by pose and shape from a 2D clothing image. Along this line, we first utilize Kinematics-aware Pose Estimation to estimate body pose parameters. 3D skeleton is employed as a proxy followed by an inverse kinematics module to boost the estimation accuracy. We additionally design an adaptive depth trick to align the re-projected 3D mesh better with 2D clothing image by disentangling the effects of object size and camera extrinsic. Next, we propose Physics-informed Shape Estimation to estimate body shape parameters. 3D shape parameters are predicted based on partial body measurements estimated from RGB image, which not only improves pixel-wise human-cloth alignment, but also enables flexible user editing. Finally, we design Evolution-based pose generation method, a skeleton transplanting method inspired by genetic algorithms to generate diverse reasonable poses during inference. As shown by experimental results on both synthetic and real-world data, the proposed framework achieves state-of-the-art performance and can effectively recover natural and diverse 3D body meshes from 2D images that align well with clothing.

cs.CV↗

arXiv2023

Decision Fusion Network with Perception Fine-tuning for Defect Classification

Surface defect inspection is an important task in industrial inspection. Deep learning-based methods have demonstrated promising performance in this domain. Nevertheless, these methods still suffer from misjudgment when encountering challenges such as low-contrast defects and complex backgrounds. To overcome these issues, we present a decision fusion network (DFNet) that incorporates the semantic decision with the feature decision to strengthen the decision ability of the network. In particular, we introduce a decision fusion module (DFM) that extracts a semantic vector from the semantic decision branch and a feature vector for the feature decision branch and fuses them to make the final classification decision. In addition, we propose a perception fine-tuning module (PFM) that fine-tunes the foreground and background during the segmentation stage. PFM generates the semantic and feature outputs that are sent to the classification decision stage. Furthermore, we present an inner-outer separation weight matrix to address the impact of label edge uncertainty during segmentation supervision. Our experimental results on the publicly available datasets including KolektorSDD2 (96.1% AP) and Magnetic-tile-defect-datasets (94.6% mAP) demonstrate the effectiveness of the proposed method.

cs.CV↗

arXiv2023

CINFormer: Transformer network with multi-stage CNN feature injection for surface defect segmentation

Surface defect inspection is of great importance for industrial manufacture and production. Though defect inspection methods based on deep learning have made significant progress, there are still some challenges for these methods, such as indistinguishable weak defects and defect-like interference in the background. To address these issues, we propose a transformer network with multi-stage CNN (Convolutional Neural Network) feature injection for surface defect segmentation, which is a UNet-like structure named CINFormer. CINFormer presents a simple yet effective feature integration mechanism that injects the multi-level CNN features of the input image into different stages of the transformer network in the encoder. This can maintain the merit of CNN capturing detailed features and that of transformer depressing noises in the background, which facilitates accurate defect detection. In addition, CINFormer presents a Top-K self-attention module to focus on tokens with more important information about the defects, so as to further reduce the impact of the redundant background. Extensive experiments conducted on the surface defect datasets DAGM 2007, Magnetic tile, and NEU show that the proposed CINFormer achieves state-of-the-art performance in defect detection.

cs.CV↗

arXiv2023

Graph Contrastive Learning Meets Graph Meta Learning: A Unified Method for Few-shot Node Tasks

Graph Neural Networks (GNNs) have become popular in Graph Representation Learning (GRL). One fundamental application is few-shot node classification. Most existing methods follow the meta learning paradigm, showing the ability of fast generalization to few-shot tasks. However, recent works indicate that graph contrastive learning combined with fine-tuning can significantly outperform meta learning methods. Despite the empirical success, there is limited understanding of the reasons behind it. In our study, we first identify two crucial advantages of contrastive learning compared to meta learning, including (1) the comprehensive utilization of graph nodes and (2) the power of graph augmentations. To integrate the strength of both contrastive learning and meta learning on the few-shot node classification tasks, we introduce a new paradigm: Contrastive Few-Shot Node Classification (COLA). Specifically, COLA employs graph augmentations to identify semantically similar nodes, which enables the construction of meta-tasks without the need for label information. Therefore, COLA can utilize all nodes to construct meta-tasks, further reducing the risk of overfitting. Through extensive experiments, we validate the essentiality of each component in our design and demonstrate that COLA achieves new state-of-the-art on all tasks.

cs.LG↗

arXiv2023

PottsMGNet: A Mathematical Explanation of Encoder-Decoder Based Neural Networks

For problems in image processing and many other fields, a large class of effective neural networks has encoder-decoder-based architectures. Although these networks have made impressive performances, mathematical explanations of their architectures are still underdeveloped. In this paper, we study the encoder-decoder-based network architecture from the algorithmic perspective and provide a mathematical explanation. We use the two-phase Potts model for image segmentation as an example for our explanations. We associate the segmentation problem with a control problem in the continuous setting. Then, multigrid method and operator splitting scheme, the PottsMGNet, are used to discretize the continuous control model. We show that the resulting discrete PottsMGNet is equivalent to an encoder-decoder-based network. With minor modifications, it is shown that a number of the popular encoder-decoder-based neural networks are just instances of the proposed PottsMGNet. By incorporating the Soft-Threshold-Dynamics into the PottsMGNet as a regularizer, the PottsMGNet has shown to be robust with the network parameters such as network width and depth and achieved remarkable performance on datasets with very large noise. In nearly all our experiments, the new network always performs better or as good on accuracy and dice score than existing networks for image segmentation.

cs.CV↗

arXiv2023

Time consistency of dynamic risk measures and dynamic performance measures generated by distortion functions

The aim of this work is to study risk measures generated by distortion functions in a dynamic discrete time setup, and to investigate the corresponding dynamic coherent acceptability indices (DCAIs) generated by families of such risk measures. First we show that conditional version of Choquet integrals indeed are dynamic coherent risk measures (DCRMs), and also introduce the class of dynamic weighted value at risk measures. We prove that these two classes of risk measures coincides. In the spirit of robust representations theorem for DCAIs, we establish some relevant properties of families of DCRMs generated by distortion functions, and then define and study the corresponding DCAIs. Second, we study the time consistency of DCRMs and DCAIs generated by distortion functions. In particular, we prove that such DCRMs are sub-martingale time consistent, but they are not super-martingale time consistent. We also show that DCRMs generated by distortion functions are not weakly acceptance time consistent. We also present several widely used classes of distortion functions and derive some new representations of these distortions.

q-fin.RM↗

arXiv2023

Attention Where It Matters: Rethinking Visual Document Understanding with Selective Region Concentration

We propose a novel end-to-end document understanding model called SeRum (SElective Region Understanding Model) for extracting meaningful information from document images, including document analysis, retrieval, and office automation. Unlike state-of-the-art approaches that rely on multi-stage technical schemes and are computationally expensive, SeRum converts document image understanding and recognition tasks into a local decoding process of the visual tokens of interest, using a content-aware token merge module. This mechanism enables the model to pay more attention to regions of interest generated by the query decoder, improving the model's effectiveness and speeding up the decoding speed of the generative scheme. We also designed several pre-training tasks to enhance the understanding and local awareness of the model. Experimental results demonstrate that SeRum achieves state-of-the-art performance on document understanding tasks and competitive results on text spotting tasks. SeRum represents a substantial advancement towards enabling efficient and effective end-to-end document understanding.

cs.CV↗

arXiv2023

On Degeneracy Issues in Multi-parametric Programming and Critical Region Exploration based Distributed Optimization in Smart Grid Operations

Improving renewable energy resource utilization efficiency is crucial to reducing carbon emissions, and multi-parametric programming has provided a systematic perspective in conducting analysis and optimization toward this goal in smart grid operations. This paper focuses on two aspects of interest related to multi-parametric linear/quadratic programming (mpLP/QP). First, we study degeneracy issues of mpLP/QP. A novel approach to deal with degeneracies is proposed to find all critical regions containing the given parameter. Our method leverages properties of the multi-parametric linear complementary problem, vertex searching technique, and complementary basis enumeration. Second, an improved critical region exploration (CRE) method to solve distributed LP/QP is proposed under a general mpLP/QP-based formulation. The improved CRE incorporates the proposed approach to handle degeneracies. A cutting plane update and an adaptive stepsize scheme are also integrated to accelerate convergence under different problem settings. The computational efficiency is verified on multi-area tie-line scheduling problems with various testing benchmarks and initial states.

eess.SY↗

arXiv2023

Rigidity of Steady Solutions to the Navier-Stokes Equations in High Dimensions and its applications

Solutions with scaling-invariant bounds such as self-similar solutions, play an important role in the understanding of the regularity and asymptotic structures of solutions to the Navier-Stokes equations. In this paper, we prove that any steady solution satisfying $|\Bu(x)|\leq C/|x|$ for any constant $C$ in $\mathbb{R}^n\setminus \{0\}$ with $ n \geq 4$, must be zero. Our main idea is to analyze the velocity field and the total head pressure via weighted energy estimates with suitable multipliers so that the proof is pretty elementary and short. These results not only give the Liouville-type theorem for steady solutions in higher dimensions with neither smallness nor self-similarity type assumptions, but also help to remove a class of singularities of solutions and give the optimal asymptotic behaviors of solutions at infinity in the exterior domains.

math.AP↗

arXiv2023

Blockwise Parallel Transformer for Large Context Models

Transformers have emerged as the cornerstone of state-of-the-art natural language processing models, showcasing exceptional performance across a wide range of AI applications. However, the memory demands posed by the self-attention mechanism and the large feedforward network in Transformers limit their ability to handle long sequences, thereby creating challenges for tasks involving multiple long sequences or long-term dependencies. We present a distinct approach, Blockwise Parallel Transformer (BPT), that leverages blockwise computation of self-attention and feedforward network fusion to minimize memory costs. By processing longer input sequences while maintaining memory efficiency, BPT enables training sequences 32 times longer than vanilla Transformers and up to 4 times longer than previous memory-efficient methods. Extensive experiments on language modeling and reinforcement learning tasks demonstrate the effectiveness of BPT in reducing memory requirements and improving performance.

cs.CL↗