arXiv subjects

Hong Liu

Publications and source records attributed to Hong Liu.

At least 235 records · Page 13Linked to original sources

arXiv2023

Uniform Sequence Better: Time Interval Aware Data Augmentation for Sequential Recommendation

Sequential recommendation is an important task to predict the next-item to access based on a sequence of interacted items. Most existing works learn user preference as the transition pattern from the previous item to the next one, ignoring the time interval between these two items. However, we observe that the time interval in a sequence may vary significantly different, and thus result in the ineffectiveness of user modeling due to the issue of \emph{preference drift}. In fact, we conducted an empirical study to validate this observation, and found that a sequence with uniformly distributed time interval (denoted as uniform sequence) is more beneficial for performance improvement than that with greatly varying time interval. Therefore, we propose to augment sequence data from the perspective of time interval, which is not studied in the literature. Specifically, we design five operators (Ti-Crop, Ti-Reorder, Ti-Mask, Ti-Substitute, Ti-Insert) to transform the original non-uniform sequence to uniform sequence with the consideration of variance of time intervals. Then, we devise a control strategy to execute data augmentation on item sequences in different lengths. Finally, we implement these improvements on a state-of-the-art model CoSeRec and validate our approach on four real datasets. The experimental results show that our approach reaches significantly better performance than the other 11 competing methods. Our implementation is available: https://github.com/KingGugu/TiCoSeRec.

cs.IR↗

arXiv2023

Vanishing codegree Turán density implies vanishing uniform Turán density

For a $k$-uniform hypergraph (or simply $k$-graph) $F$, the codegree Turán density $π_{\mathrm{co}}(F)$ is the infimum over all $α$ such that any $n$-vertex $k$-graph $H$ with every $(k-1)$-subset of $V(H)$ contained in at least $αn$ edges has a copy of $F$. The uniform Turán density $π_{\therefore}(F)$ is the supremum over all $d$ such that there are infinitely many $F$-free $k$-graphs $H$ satisfying that any linear-size subhypergraph of $H$ has edge density at least $d$. Falgas-Ravry, Pikhurko, Vaughan and Volec [J. London Math. Soc., 2023] asked whether for every $3$-graph $F$, $π_{\therefore}(F)\leqπ_{\mathrm{co}}(F)$. We provide a positive answer to this question provided that $π_{\mathrm{co}}(F)=0$. Our proof relies on a random geometric construction and a new formulation of the characterization of $3$-graphs with vanishing uniform Turán density due to Reiher, R{ö}dl and Schacht [J. London Math. Soc., 2018]. Along the way, we answer a question of Falgas-Ravry, Pikhurko, Vaughan and Volec about subhypergraphs with linear minimum codegree in uniformly dense hypergraphs in the negative.

math.CO↗

arXiv2023

Simultaneous Alignment and Surface Regression Using Hybrid 2D-3D Networks for 3D Coherent Layer Segmentation of Retinal OCT Images with Full and Sparse Annotations

Layer segmentation is important to quantitative analysis of retinal optical coherence tomography (OCT). Recently, deep learning based methods have been developed to automate this task and yield remarkable performance. However, due to the large spatial gap and potential mismatch between the B-scans of an OCT volume, all of them were based on 2D segmentation of individual B-scans, which may lose the continuity and diagnostic information of the retinal layers in 3D space. Besides, most of these methods required dense annotation of the OCT volumes, which is labor-intensive and expertise-demanding. This work presents a novel framework based on hybrid 2D-3D convolutional neural networks (CNNs) to obtain continuous 3D retinal layer surfaces from OCT volumes, which works well with both full and sparse annotations. The 2D features of individual B-scans are extracted by an encoder consisting of 2D convolutions. These 2D features are then used to produce the alignment displacement vectors and layer segmentation by two 3D decoders coupled via a spatial transformer module. Two losses are proposed to utilize the retinal layers' natural property of being smooth for B-scan alignment and layer segmentation, respectively, and are the key to the semi-supervised learning with sparse annotation. The entire framework is trained end-to-end. To the best of our knowledge, this is the first work that attempts 3D retinal layer segmentation in volumetric OCT images based on CNNs. Experiments on a synthetic dataset and three public clinical datasets show that our framework can effectively align the B-scans for potential motion correction, and achieves superior performance to state-of-the-art 2D deep learning methods in terms of both layer segmentation accuracy and cross-B-scan 3D continuity in both fully and semi-supervised settings, thus offering more clinical values than previous works.

eess.IV↗

arXiv2023

Revisiting and Advancing Adversarial Training Through A Simple Baseline

In this paper, we delve into the essential components of adversarial training which is a pioneering defense technique against adversarial attacks. We indicate that some factors such as the loss function, learning rate scheduler, and data augmentation, which are independent of the model architecture, will influence adversarial robustness and generalization. When these factors are controlled for, we introduce a simple baseline approach, termed SimpleAT, that performs competitively with recent methods and mitigates robust overfitting. We conduct extensive experiments on CIFAR-10/100 and Tiny-ImageNet, which validate the robustness of SimpleAT against state-of-the-art adversarial attackers such as AutoAttack. Our results also demonstrate that SimpleAT exhibits good performance in the presence of various image corruptions, such as those found in the CIFAR-10-C. In addition, we empirically show that SimpleAT is capable of reducing the variance in model predictions, which is considered the primary contributor to robust overfitting. Our results also reveal the connections between SimpleAT and many advanced state-of-the-art adversarial defense methods.

cs.CV↗

arXiv2023

Semantic-aware Consistency Network for Cloth-changing Person Re-Identification

Cloth-changing Person Re-Identification (CC-ReID) is a challenging task that aims to retrieve the target person across multiple surveillance cameras when clothing changes might happen. Despite recent progress in CC-ReID, existing approaches are still hindered by the interference of clothing variations since they lack effective constraints to keep the model consistently focused on clothing-irrelevant regions. To address this issue, we present a Semantic-aware Consistency Network (SCNet) to learn identity-related semantic features by proposing effective consistency constraints. Specifically, we generate the black-clothing image by erasing pixels in the clothing area, which explicitly mitigates the interference from clothing variations. In addition, to fully exploit the fine-grained identity information, a head-enhanced attention module is introduced, which learns soft attention maps by utilizing the proposed part-based matching loss to highlight head information. We further design a semantic consistency loss to facilitate the learning of high-level identity-related semantic features, forcing the model to focus on semantically consistent cloth-irrelevant regions. By using the consistency constraint, our model does not require any extra auxiliary segmentation module to generate the black-clothing image or locate the head region during the inference stage. Extensive experiments on four cloth-changing person Re-ID datasets (LTCC, PRCC, Vc-Clothes, and DeepChange) demonstrate that our proposed SCNet makes significant improvements over prior state-of-the-art approaches. Our code is available at: https://github.com/Gpn-star/SCNet.

cs.CV↗

arXiv2023

Prompt Pool based Class-Incremental Continual Learning for Dialog State Tracking

Continual learning is crucial for dialog state tracking (DST) in dialog systems, since requirements from users for new functionalities are often encountered. However, most of existing continual learning methods for DST require task identities during testing, which is a severe limit in real-world applications. In this paper, we aim to address continual learning of DST in the class-incremental scenario (namely the task identity is unknown in testing). Inspired by the recently emerging prompt tuning method that performs well on dialog systems, we propose to use the prompt pool method, where we maintain a pool of key-value paired prompts and select prompts from the pool according to the distance between the dialog history and the prompt keys. The proposed method can automatically identify tasks and select appropriate prompts during testing. We conduct experiments on Schema-Guided Dialog dataset (SGD) and another dataset collected from a real-world dialog application. Experiment results show that the prompt pool method achieves much higher joint goal accuracy than the baseline. After combining with a rehearsal buffer, the model performance can be further improved.

cs.CL↗

arXiv2023

Causal connectability between quantum systems and the black hole interior in holographic duality

In holographic duality an eternal AdS black hole is described by two copies of the boundary CFT in the thermal field double state. This identification has many puzzles, including the boundary descriptions of the event horizons, the interiors of the black hole, and the singularities. Compounding these mysteries is the fact that, while there is no interaction between the CFTs, observers from them can fall into the black hole and interact. We address these issues in this paper. In particular, we (i) present a boundary formulation of a class of in-falling bulk observers; (ii) present an argument that a sharp bulk event horizon can only emerge in the infinite $N$ limit of the boundary theory; (iii) give an explicit construction in the boundary theory of an evolution operator for a bulk in-falling observer, making manifest the boundary emergence of the black hole horizons, the interiors, and the associated causal structure. A by-product is a concept called causal connectability, which is a criterion for any two quantum systems (which do not need to have a known gravity dual) to have an emergent sharp horizon structure.

hep-th↗

arXiv2023

Emergent times in holographic duality

In holographic duality an eternal AdS black hole is described by two copies of the boundary CFT in the thermal field double state. In this paper we provide explicit constructions in the boundary theory of infalling time evolutions which can take bulk observers behind the horizon. The constructions also help to illuminate the boundary emergence of the black hole horizons, the interiors, and the associated causal structure. A key element is the emergence, in the large $N$ limit of the boundary theory, of a type III$_1$ von Neumann algebraic structure from the type I boundary operator algebra and the half-sided modular translation structure associated with it.

hep-th↗

arXiv2023

Stability from graph symmetrisation arguments with applications to inducibility

We present a sufficient condition for the stability property of extremal graph problems that can be solved via Zykov's symmetrisation. Our criterion is stated in terms of an analytic limit version of the problem. We show that, for example, it applies to the inducibility problem for an arbitrary complete bipartite graph $B$, which asks for the maximum number of induced copies of $B$ in an $n$-vertex graph, and to the inducibility problem for $K_{2,1,1,1}$ and $K_{3,1,1}$, the only complete partite graphs on at most five vertices for which the problem was previously open.

math.CO↗

arXiv2023

Semi-supervised Sound Event Detection with Local and Global Consistency Regularization

Learning meaningful frame-wise features on a partially labeled dataset is crucial to semi-supervised sound event detection. Prior works either maintain consistency on frame-level predictions or seek feature-level similarity among neighboring frames, which cannot exploit the potential of unlabeled data. In this work, we design a Local and Global Consistency (LGC) regularization scheme to enhance the model on both label- and feature-level. The audio CutMix is introduced to change the contextual information of clips. Then, the local consistency is adopted to encourage the model to leverage local features for frame-level predictions, and the global consistency is applied to force features to align with global prototypes through a specially designed contrastive loss. Experiments on the DESED dataset indicate the superiority of LGC, surpassing its respective competitors largely with the same settings as the baseline system. Besides, combining LGC with existing methods can obtain further improvements. The code will be released soon.

eess.AS↗

arXiv2023

Audio-free Prompt Tuning for Language-Audio Models

Contrastive Language-Audio Pretraining (CLAP) is pre-trained to associate audio features with human language, making it a natural zero-shot classifier to recognize unseen sound categories. To adapt CLAP to downstream tasks, prior works inevitably require labeled domain audios, which limits their scalability under data scarcity and deprives them of the capability to detect novel classes as the original CLAP. In this work, by leveraging the modality alignment in CLAP, we propose an efficient audio-free prompt tuning scheme aimed at optimizing a few prompt tokens from texts instead of audios, which regularizes the model space to avoid overfitting the seen classes as well. Based on this, a multi-grained prompt design is further explored to fuse global and local information. Experiments on several tasks demonstrate that our approach can boost the CLAP and outperform other training methods on model performance and training efficiency. While conducting zero-shot inference on unseen categories, it still shows better transferability than the vanilla CLAP. Moreover, our method is flexible enough even if only knowing the downstream class names. The code will be released soon.

eess.AS↗

arXiv2023

An Efficient Trajectory Planner for Car-like Robots on Uneven Terrain

Autonomous navigation of ground robots on uneven terrain is being considered in more and more tasks. However, uneven terrain will bring two problems to motion planning: how to assess the traversability of the terrain and how to cope with the dynamics model of the robot associated with the terrain. The trajectories generated by existing methods are often too conservative or cannot be tracked well by the controller since the second problem is not well solved. In this paper, we propose terrain pose mapping to describe the impact of terrain on the robot. With this mapping, we can obtain the SE(3) state of the robot on uneven terrain for a given state in SE(2). Then, based on it, we present a trajectory optimization framework for car-like robots on uneven terrain that can consider both of the above problems. The trajectories generated by our method conform to the dynamics model of the system without being overly conservative and yet able to be tracked well by the controller. We perform simulations and real-world experiments to validate the efficiency and trajectory quality of our algorithm.

cs.RO↗

arXiv2023

Studies of Nonadiabatic Dynamics in the Singlet Fission Processes of Pentacene Dimer via Tensor Train Decomposition Method

Singlet fission (SF) is a very significant photophysical phenomenon and possesses potential applications. In this work, we try to give the rather detailed theoretical investigation of the SF process in the stacked polyacene dimer by combining the high-level quantum chemistry calculations, and the quantum dynamics simulations based on the tensor train decomposition method. Starting from the construction of the linear vibronic coupling model, we explore the pure electronic dynamics and the vibronic dynamics in the SF processes. The role of vibrational modes in nonadiabatic dynamics is addressed. The results show that the super-exchange mechanism mediated by the charge-transfer state is found in both pure electronic dynamics and the nonadiabatic dynamics. Particularly, the vibrational modes with the frequency resonance with the adiabatic energy gap play very import roles in the SF dynamics. This work not only provides a deep and detailed understanding of the SF process, but also verifies the efficiency of the tensor train decomposition method that can serve as the reference dynamics method to explore the dynamics behaviors of complex systems.

physics.chem-ph↗

arXiv2023

Extremal number of graphs from geometric shapes

We study the Turán problem for highly symmetric bipartite graphs arising from geometric shapes and periodic tilings commonly found in nature. 1. The prism $C_{2\ell}^{\square}:=C_{2\ell}\square K_{2}$ is the graph consisting of two vertex disjoint $2\ell$-cycles and a matching pairing the corresponding vertices of these two cycles. We show that for every $\ell\ge 4$, ex$(n,C_{2\ell}^{\square})=Θ(n^{3/2})$. This resolves a conjecture of He, Li and Feng. 2. The hexagonal tiling in honeycomb is one of the most natural structures in the real world. We show that the extremal number of honeycomb graphs has the same order of magnitude as their basic building unit 6-cycles. 3. We also consider bipartite graphs from quadrangulations of the cylinder and the torus. We prove near optimal bounds for both configurations. In particular, our method gives a very short proof of a tight upper bound for the extremal number of the 2-dimensional grid, improving a recent result of Bradač, Janzer, Sudakov and Tomon. Our proofs mix several ideas, including shifting embedding schemes, weighted homomorphism and subgraph counts and asymmetric dependent random choice.

math.CO↗

arXiv2023

Crux, space constraints and subdivisions

For a given graph $H$, its subdivisions carry the same topological structure. The existence of $H$-subdivisions within a graph $G$ has deep connections with topological, structural and extremal properties of $G$. One prominent example of such a connection, due to Bollobás and Thomason and independently Komlós and Szemerédi, asserts that the average degree of $G$ being $d$ ensures a $K_{Ω(\sqrt{d})}$-subdivision in $G$. Although this square-root bound is best possible, various results showed that much larger clique subdivisions can be found in a graph for many natural classes. We investigate the connection between crux, a notion capturing the essential order of a graph, and the existence of large clique subdivisions. This reveals the unifying cause underpinning all those improvements for various classes of graphs studied. Roughly speaking, when embedding subdivisions, natural space constraints arise; and such space constraints can be measured via crux. Our main result gives an asymptotically optimal bound on the size of a largest clique subdivision in a generic graph $G$, which is determined by both its average degree and its crux size. As corollaries, we obtain (1) a characterisation of extremal graphs for which the square-root bound above is tight: they are essentially disjoint unions of graphs having crux size linear in $d$; (2) a unifying approach to find a clique subdivision of almost optimal size in graphs which do not contain a fixed bipartite graph as a subgraph; (3) and that the clique subdivision size in random graphs $G(n,p)$ witnesses a dichotomy: when $p = ω(n^{-1/2})$, the barrier is the space, while when $p=o( n^{-1/2})$, the bottleneck is the density.

math.CO↗

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

An effective field theory for non-maximal quantum chaos

In non-maximally quantum chaotic systems, the exponential behavior of out-of-time-ordered correlators (OTOCs) results from summing over exchanges of an infinite tower of higher "spin" operators. We construct an effective field theory (EFT) to capture these exchanges in $(0+1)$ dimensions. The EFT generalizes the one for maximally chaotic systems, and reduces to it in the limit of maximal chaos. The theory predicts the general structure of OTOCs both at leading order in the $1/N$ expansion ($N$ is the number of degrees of freedom), and after resuming over an infinite number of higher order $1/N$ corrections. These general results agree with those previously explicitly obtained in specific models. We also show that the general structure of the EFT can be extracted from the large $q$ SYK model.

hep-th↗

arXiv2023

The Erdős distinct subset sums problem in a modular setting

We prove the following variant of the Erdős distinct subset sums problem. Given $t \ge 0$ and sufficiently large $n$, every $n$-element set $A$ whose subset sums are distinct modulo $N=2^n+t$ satisfies $$\max A \ge \Big(\frac{1}{3}-o(1)\Big)N.$$ Furthermore, we provide examples showing that the constant $\frac 13$ is best possible. For small values of $t$, we characterise the structure of all sumset-distinct sets modulo $N=2^n+t$ of cardinality $n$.

math.CO↗