arXiv subjects

Yunpeng Liu

Publications and source records attributed to Yunpeng Liu.

At least 55 records · Page 3Linked to original sources

arXiv2023

A Diffusion-Model of Joint Interactive Navigation

Simulation of autonomous vehicle systems requires that simulated traffic participants exhibit diverse and realistic behaviors. The use of prerecorded real-world traffic scenarios in simulation ensures realism but the rarity of safety critical events makes large scale collection of driving scenarios expensive. In this paper, we present DJINN - a diffusion based method of generating traffic scenarios. Our approach jointly diffuses the trajectories of all agents, conditioned on a flexible set of state observations from the past, present, or future. On popular trajectory forecasting datasets, we report state of the art performance on joint trajectory metrics. In addition, we demonstrate how DJINN flexibly enables direct test-time sampling from a variety of valuable conditional distributions including goal-based sampling, behavior-class sampling, and scenario editing.

cs.LG↗

arXiv2023

Self-supervision meets kernel graph neural models: From architecture to augmentations

Graph representation learning has now become the de facto standard when handling graph-structured data, with the framework of message-passing graph neural networks (MPNN) being the most prevailing algorithmic tool. Despite its popularity, the family of MPNNs suffers from several drawbacks such as transparency and expressivity. Recently, the idea of designing neural models on graphs using the theory of graph kernels has emerged as a more transparent as well as sometimes more expressive alternative to MPNNs known as kernel graph neural networks (KGNNs). Developments on KGNNs are currently a nascent field of research, leaving several challenges from algorithmic design and adaptation to other learning paradigms such as self-supervised learning. In this paper, we improve the design and learning of KGNNs. Firstly, we extend the algorithmic formulation of KGNNs by allowing a more flexible graph-level similarity definition that encompasses former proposals like random walk graph kernel, as well as providing a smoother optimization objective that alleviates the need of introducing combinatorial learning procedures. Secondly, we enhance KGNNs through the lens of self-supervision via developing a novel structure-preserving graph data augmentation method called latent graph augmentation (LGA). Finally, we perform extensive empirical evaluations to demonstrate the efficacy of our proposed mechanisms. Experimental results over benchmark datasets suggest that our proposed model achieves competitive performance that is comparable to or sometimes outperforming state-of-the-art graph representation learning frameworks with or without self-supervision on graph classification tasks. Comparisons against other previously established graph data augmentation methods verify that the proposed LGA augmentation scheme captures better semantics of graph-level invariance.

cs.LG↗

arXiv2023

Video Killed the HD-Map: Predicting Multi-Agent Behavior Directly From Aerial Images

The development of algorithms that learn multi-agent behavioral models using human demonstrations has led to increasingly realistic simulations in the field of autonomous driving. In general, such models learn to jointly predict trajectories for all controlled agents by exploiting road context information such as drivable lanes obtained from manually annotated high-definition (HD) maps. Recent studies show that these models can greatly benefit from increasing the amount of human data available for training. However, the manual annotation of HD maps which is necessary for every new location puts a bottleneck on efficiently scaling up human traffic datasets. We propose an aerial image-based map (AIM) representation that requires minimal annotation and provides rich road context information for traffic agents like pedestrians and vehicles. We evaluate multi-agent trajectory prediction using the AIM by incorporating it into a differentiable driving simulator as an image-texture-based differentiable rendering module. Our results demonstrate competitive multi-agent trajectory prediction performance especially for pedestrians in the scene when using our AIM representation as compared to models trained with rasterized HD maps.

cs.CV↗

arXiv2023

Realistically distributing object placements in synthetic training data improves the performance of vision-based object detection models

When training object detection models on synthetic data, it is important to make the distribution of synthetic data as close as possible to the distribution of real data. We investigate specifically the impact of object placement distribution, keeping all other aspects of synthetic data fixed. Our experiment, training a 3D vehicle detection model in CARLA and testing on KITTI, demonstrates a substantial improvement resulting from improving the object placement distribution.

cs.CV↗

arXiv2023

Critic Sequential Monte Carlo

We introduce CriticSMC, a new algorithm for planning as inference built from a composition of sequential Monte Carlo with learned Soft-Q function heuristic factors. These heuristic factors, obtained from parametric approximations of the marginal likelihood ahead, more effectively guide SMC towards the desired target distribution, which is particularly helpful for planning in environments with hard constraints placed sparsely in time. Compared with previous work, we modify the placement of such heuristic factors, which allows us to cheaply propose and evaluate large numbers of putative action particles, greatly increasing inference and planning efficiency. CriticSMC is compatible with informative priors, whose density function need not be known, and can be used as a model-free control algorithm. Our experiments on collision avoidance in a high-dimensional simulated driving task show that CriticSMC significantly reduces collision rates at a low computational cost while maintaining realism and diversity of driving behaviors across vehicles and environment scenarios.

stat.ML↗

arXiv2022

Relativistic correction to the dissociation temperature of B_c mesons in the hot medium

By solving two body Dirac equations with potentials at finite temperature, we calculated the dissociation temperature T_d of B_c mesons in the quark-gluon plasma. It is found that the T_d becomes higher with the relativistic correction than the T_d from the Schroedinger equation. Both the short range interaction and the constant term of the potential at the long-range scale have a contribution to the shift of T_d, while the spin interaction is negligible.

nucl-th↗

arXiv2022

OBBStacking: An Ensemble Method for Remote Sensing Object Detection

Ensemble methods are a reliable way to combine several models to achieve superior performance. However, research on the application of ensemble methods in the remote sensing object detection scenario is mostly overlooked. Two problems arise. First, one unique characteristic of remote sensing object detection is the Oriented Bounding Boxes (OBB) of the objects and the fusion of multiple OBBs requires further research attention. Second, the widely used deep learning object detectors provide a score for each detected object as an indicator of confidence, but how to use these indicators effectively in an ensemble method remains a problem. Trying to address these problems, this paper proposes OBBStacking, an ensemble method that is compatible with OBBs and combines the detection results in a learned fashion. This ensemble method helps take 1st place in the Challenge Track \textit{Fine-grained Object Recognition in High-Resolution Optical Images}, which was featured in \textit{2021 Gaofen Challenge on Automated High-Resolution Earth Observation Image Interpretation}. The experiments on DOTA dataset and FAIR1M dataset demonstrate the improved performance of OBBStacking and the features of OBBStacking are analyzed.

cs.CV↗

arXiv2022

Vehicle Type Specific Waypoint Generation

We develop a generic mechanism for generating vehicle-type specific sequences of waypoints from a probabilistic foundation model of driving behavior. Many foundation behavior models are trained on data that does not include vehicle information, which limits their utility in downstream applications such as planning. Our novel methodology conditionally specializes such a behavior predictive model to a vehicle-type by utilizing byproducts of the reinforcement learning algorithms used to produce vehicle specific controllers. We show how to compose a vehicle specific value function estimate with a generic probabilistic behavior model to generate vehicle-type specific waypoint sequences that are more likely to be physically plausible then their vehicle-agnostic counterparts.

cs.AI↗

arXiv2022

$B_c^+$ Formation from Random Charm and Anti-bottom Quarks in the Quark-Gluon Plasma

We study the $B_c^+$ production in Pb-Pb collisions at $\sqrt{s_{NN}}=5.02$ TeV. In the quark-gluon plasma (QGP) produced in heavy-ion collisions, heavy quarks make random motions with the energy loss. We employ the Langevin equations to study the non-equilibrium distributions of heavy quarks and the Instantaneous Coalescence Model (ICM) to study the hadronization process. Due to abundant charm and bottom quarks in the QGP, their coalescence probability is significantly enhanced compared with the situations in proton-proton collisions. We find that the final production of $B_c^+$ is increased by the coalescence process, which makes the nuclear modification factor ($R_{AA}$) of $B_c^+$ larger than the unit. Our model explains the experimental data well at semi-central and central collisions. The observation of $R_{AA}(B_c^+)>1$ is regarded as an evident and strong signal of the existence of the deconfined medium generated in heavy-ion collisions.

nucl-th↗

arXiv2022

X(3872) Production in Relativistic Heavy-Ion Collisions

Heavy-ion collisions provide a unique opportunity to study the nature of X(3872) compared with electron-positron and proton-proton (antiproton) collisions. We investigate the centrality and momentum dependence of X(3872) in heavy-ion collisions via the Langevin equation and instant coalescence model (LICM). When X(3872) is treated as a compact tetraquark state, the tetraquarks are produced via the coalescence of heavy and light quarks near the quantum chromodynamic (QCD) phase transition due to the restoration of the heavy quark potential at $T\rightarrow T_c$. In the molecular scenario, loosely bound X(3872) is produced via the coalescence of $D^0$-$\bar D^{*0}$ mesons in a hadronic medium after kinetic freeze-out. We employ the LICM to explain both $D^0$ and $J/ψ$ production as a benchmark. Then we give predictions regarding X(3872) production and the nuclear modification factor $R_{AA}^{X(3872)}$. We find that the total yield of tetraquark is several times larger than the molecular production in Pb-Pb collisions. Although the geometric size of the molecule is huge, the coalescence probability is small due to strict constraints on the relative momentum between $D^0$ and $\bar D^{*0}$ in the molecular Wigner function, which significantly suppresses the molecular yield.

hep-ph↗

arXiv2022

Conditional Permutation Invariant Flows

We present a novel, conditional generative probabilistic model of set-valued data with a tractable log density. This model is a continuous normalizing flow governed by permutation equivariant dynamics. These dynamics are driven by a learnable per-set-element term and pairwise interactions, both parametrized by deep neural networks. We illustrate the utility of this model via applications including (1) complex traffic scene generation conditioned on visually specified map information, and (2) object bounding box generation conditioned directly on images. We train our model by maximizing the expected likelihood of labeled conditional data under our flow, with the aid of a penalty that ensures the dynamics are smooth and hence efficiently solvable. Our method significantly outperforms non-permutation invariant baselines in terms of log likelihood and domain-specific metrics (offroad, collision, and combined infractions), yielding realistic samples that are difficult to distinguish from real data.

stat.ML↗

arXiv2022

Multiplication of freestanding semiconductor membranes from a single wafer by advanced remote epitaxy

Freestanding single-crystalline membranes are an important building block for functional electronics. Especially, compounds semiconductor membranes such as III-N and III-V offer great opportunities for optoelectronics, high-power electronics, and high-speed computing. Despite huge efforts to produce such membranes by detaching epitaxial layers from donor wafers, however, it is still challenging to harvest epitaxial layers using practical processes. Here, we demonstrate a method to grow and harvest multiple epitaxial membranes with extremely high throughput at the wafer scale. For this, 2D materials are directly formed on III-N and III-V substrates in epitaxy systems, which enables an advanced remote epitaxy scheme comprised of multiple alternating layers of 2D materials and epitaxial layers that can be formed by a single epitaxy run. Each epilayer in the multi-stack structure is then harvested by layer-by-layer peeling, producing multiple freestanding membranes with unprecedented throughput from a single wafer. Because 2D materials allow peeling at the interface without damaging the epilayer or the substrate, wafers can be reused for subsequent membrane production. Therefore, this work represents a meaningful step toward high-throughput and low-cost production of single-crystal membranes that can be heterointegrated.

physics.app-ph↗

arXiv2021

Hadronic effects on charmonium elliptic flows in heavy-ion collisions

Transport and Langevin equations are employed to study hadronic medium effects on charmonium elliptic flows in heavy-ion collisions. In Pb-Pb collisions, the anisotropic energy density of the quark-gluon plasma (QGP) in the transverse plane is transformed into hadron momentum anisotropy after the phase transition. Charmonia with high transverse momentum $p_T$ are produced via the primordial hard process and undergo different degrees of dissociation along different paths in the QGP. They then scatter with light hadrons in the hadron phase. Both contributions to the charmonium elliptic flows are studied at moderate and high transverse momenta. The elliptic flows of the prompt $J/ψ$ are found to be considerably enhanced at high transverse momentum when the charmonium diffusion coefficients in the hadronic medium are parametrized through the geometry scale approximation. This hadronic medium effect is negligible for quarkonia with larger mass such as bottomonia.

nucl-th↗

arXiv2020

Critical temperature of deconfinement in a constrained space using a bag model at vanishing baryon density

The geometry of fireballs in relativistic heavy ion collisions is approximated by a static box, which is infinite in two directions while finite in the other direction. The critical temperature of deconfinement phase transition is calculated explicitly in the MIT bag model at vanishing baryon density. It is found that the critical temperature shifts to a value higher than that in an unconstrained space.

nucl-th↗

arXiv2020

Compact Learning for Multi-Label Classification

Multi-label classification (MLC) studies the problem where each instance is associated with multiple relevant labels, which leads to the exponential growth of output space. MLC encourages a popular framework named label compression (LC) for capturing label dependency with dimension reduction. Nevertheless, most existing LC methods failed to consider the influence of the feature space or misguided by original problematic features, so that may result in performance degeneration. In this paper, we present a compact learning (CL) framework to embed the features and labels simultaneously and with mutual guidance. The proposal is a versatile concept, hence the embedding way is arbitrary and independent of the subsequent learning process. Following its spirit, a simple yet effective implementation called compact multi-label learning (CMLL) is proposed to learn a compact low-dimensional representation for both spaces. CMLL maximizes the dependence between the embedded spaces of the labels and features, and minimizes the loss of label space recovery concurrently. Theoretically, we provide a general analysis for different embedding methods. Practically, we conduct extensive experiments to validate the effectiveness of the proposed method.

cs.LG↗

arXiv2020

Fast Heating Dissociation of Upsilon(1S) in Heavy Ion Collisions at RHIC

With the adiabatic assumption in the cooling process, we discussed a new mechanism on Upsilon(1S) suppression that is due to the fast heating process at the early stage of the fireball instead of its finite decay width in finite temperature medium produced in the heavy ion collisions. We calculated the transition probability after the fast heating dissociation as a function of the temperature of the medium and the nuclear modification factor in central collisions, and found that the suppression is not negligible at RHIC, even if the width of Upsilon(1S) vanishes.

nucl-th↗

arXiv2020

A modified Euler-Maclaurin formula in 1D and 2D with applications in statistical physics

The Euler-Maclaurin summation formula is generalized to a modified form by expanding the periodic Bernoulli polynomials as its Fourier series and taking cuts, which includes both the Euler-Maclaurin summation formula and the Poission summation formula as special cases. By making use of the modified formula, a numerical summation method is obtained and the error can be controlled. The modified formula is also generalized from one dimention to two dimentions. Examples of its applications in statistical physics are also discussed.

math-ph↗

arXiv2018

Heavy quarkonium dissociation in the finite space of heavy-ion collisions

The dissociation of heavy quarkonia in the constrained space is calculated at leading order compared with that in infinitely large medium. To deal with the summation of the discrete spectrum, a modified Euler-Maclaurin formula is developed as our numerical algorithm. We find that with the constraint in space, the dissociation of quarkonia at early time becomes negligible.

nucl-th↗