arXiv subjects

Chang Zhou

Publications and source records attributed to Chang Zhou.

At least 73 records · Page 4Linked to original sources

arXiv2024

Multi-Scenario Combination Based on Multi-Agent Reinforcement Learning to Optimize the Advertising Recommendation System

This paper explores multi-scenario optimization on large platforms using multi-agent reinforcement learning (MARL). We address this by treating scenarios like search, recommendation, and advertising as a cooperative, partially observable multi-agent decision problem. We introduce the Multi-Agent Recurrent Deterministic Policy Gradient (MARDPG) algorithm, which aligns different scenarios under a shared objective and allows for strategy communication to boost overall performance. Our results show marked improvements in metrics such as click-through rate (CTR), conversion rate, and total sales, confirming our method's efficacy in practical settings.

cs.LG↗

arXiv2024

Can Large Language Models Always Solve Easy Problems if They Can Solve Harder Ones?

Large language models (LLMs) have demonstrated impressive capabilities, but still suffer from inconsistency issues (e.g. LLMs can react differently to disturbances like rephrasing or inconsequential order change). In addition to these inconsistencies, we also observe that LLMs, while capable of solving hard problems, can paradoxically fail at easier ones. To evaluate this hard-to-easy inconsistency, we develop the ConsisEval benchmark, where each entry comprises a pair of questions with a strict order of difficulty. Furthermore, we introduce the concept of consistency score to quantitatively measure this inconsistency and analyze the potential for improvement in consistency by relative consistency score. Based on comprehensive experiments across a variety of existing models, we find: (1) GPT-4 achieves the highest consistency score of 92.2\% but is still inconsistent to specific questions due to distraction by redundant information, misinterpretation of questions, etc.; (2) models with stronger capabilities typically exhibit higher consistency, but exceptions also exist; (3) hard data enhances consistency for both fine-tuning and in-context learning. Our data and code will be publicly available on GitHub.

cs.CL↗

arXiv2024

Nuclear mass table in deformed relativistic Hartree-Bogoliubov theory in continuum, II: Even-$Z$ nuclei

The mass table in the deformed relativistic Hartree-Bogoliubov theory in continuum (DRHBc) with the PC-PK1 density functional has been established for even-$Z$ nuclei with $8\le Z\le120$, extended from the previous work for even-even nuclei [Zhang $\it{et.~al.}$ (DRHBc Mass Table Collaboration), At. Data Nucl. Data Tables 144, 101488 (2022)]. The calculated binding energies, two-nucleon and one-neutron separation energies, root-mean-square (rms) radii of neutron, proton, matter, and charge distributions, quadrupole deformations, and neutron and proton Fermi surfaces are tabulated and compared with available experimental data. A total of 4829 even-$Z$ nuclei are predicted to be bound, with an rms deviation of 1.477 MeV from the 1244 mass data. Good agreement with the available experimental odd-even mass differences, $α$ decay energies, and charge radii is also achieved. The description accuracy for nuclear masses and nucleon separation energies as well as the prediction for drip lines is compared with the results obtained from other relativistic and nonrelativistic density functional. The comparison shows that the DRHBc theory with PC-PK1 provides an excellent microscopic description for the masses of even-$Z$ nuclei. The systematics of the nucleon separation energies, odd-even mass differences, pairing energies, two-nucleon gaps, $α$ decay energies, rms radii, quadrupole deformations, potential energy curves, neutron density distributions, and neutron mean-field potentials are discussed.

nucl-th↗

arXiv2024

How Abilities in Large Language Models are Affected by Supervised Fine-tuning Data Composition

Large language models (LLMs) with enormous pre-training tokens and parameters emerge diverse abilities, including math reasoning, code generation, and instruction following. These abilities are further enhanced by supervised fine-tuning (SFT). While the open-source community has explored ad-hoc SFT for enhancing individual capabilities, proprietary LLMs exhibit versatility across various skills. Therefore, understanding the facilitation of multiple abilities via SFT is paramount. In this study, we specifically focuses on the interplay of data composition between mathematical reasoning, code generation, and general human-aligning abilities during SFT. We propose four intriguing research questions to explore the association between model performance and various factors including data amount, composition ratio, model size and SFT strategies. Our experiments reveal that distinct capabilities scale differently and larger models generally show superior performance with same amount of data. Mathematical reasoning and code generation consistently improve with increasing data amount, whereas general abilities plateau after roughly a thousand samples. Moreover, we observe data composition appears to enhance various abilities under limited data conditions, yet can lead to performance conflicts when data is plentiful. Our findings also suggest the amount of composition data influences performance more than the composition ratio. In analysis of SFT strategies, we find that sequentially learning multiple skills risks catastrophic forgetting. Our proposed Dual-stage Mixed Fine-tuning (DMT) strategy offers a promising solution to learn multiple abilities with different scaling patterns.

cs.CL↗

arXiv2024

Deep Learning Powered Estimate of The Extrinsic Parameters on Unmanned Surface Vehicles

Unmanned Surface Vehicles (USVs) are pivotal in marine exploration, but their sensors' accuracy is compromised by the dynamic marine environment. Traditional calibration methods fall short in these conditions. This paper introduces a deep learning architecture that predicts changes in the USV's dynamic metacenter and refines sensors' extrinsic parameters in real time using a Time-Sequence General Regression Neural Network (GRNN) with Euler angles as input. Simulation data from Unity3D ensures robust training and testing. Experimental results show that the Time-Sequence GRNN achieves the lowest mean squared error (MSE) loss, outperforming traditional neural networks. This method significantly enhances sensor calibration for USVs, promising improved data accuracy in challenging maritime conditions. Future work will refine the network and validate results with real-world data.

cs.RO↗

arXiv2024

Predict Click-Through Rates with Deep Interest Network Model in E-commerce Advertising

This paper proposes new methods to enhance click-through rate (CTR) prediction models using the Deep Interest Network (DIN) model, specifically applied to the advertising system of Alibaba's Taobao platform. Unlike traditional deep learning approaches, this research focuses on localized user behavior activation for tailored ad targeting by leveraging extensive user behavior data. Compared to traditional models, this method demonstrates superior ability to handle diverse and dynamic user data, thereby improving the efficiency of ad systems and increasing revenue.

cs.IR↗

arXiv2024

Research on Driver Facial Fatigue Detection Based on Yolov8 Model

In a society where traffic accidents frequently occur, fatigue driving has emerged as a grave issue. Fatigue driving detection technology, especially those based on the YOLOv8 deep learning model, has seen extensive research and application as an effective preventive measure. This paper discusses in depth the methods and technologies utilized in the YOLOv8 model to detect driver fatigue, elaborates on the current research status both domestically and internationally, and systematically introduces the processing methods and algorithm principles for various datasets. This study aims to provide a robust technical solution for preventing and detecting fatigue driving, thereby contributing significantly to reducing traffic accidents and safeguarding lives.

cs.CV↗

arXiv2024

Training-Free Long-Context Scaling of Large Language Models

The ability of Large Language Models (LLMs) to process and generate coherent text is markedly weakened when the number of input tokens exceeds their pretraining length. Given the expensive overhead of finetuning large-scale models with longer sequences, we propose Dual Chunk Attention (DCA), which enables Llama2 70B to support context windows of more than 100k tokens without continual training. By decomposing the attention computation for long sequences into chunk-based modules, DCA manages to effectively capture the relative positional information of tokens within the same chunk (Intra-Chunk) and across distinct chunks (Inter-Chunk), as well as integrates seamlessly with Flash Attention. In addition to its impressive extrapolation capability, DCA achieves performance on practical long-context tasks that is comparable to or even better than that of finetuned models. When compared with proprietary models, our training-free 70B model attains 94% of the performance of gpt-3.5-16k, indicating it is a viable open-source alternative. All code and data used in this work are released at \url{https://github.com/HKUNLP/ChunkLlama}.

cs.CL↗

arXiv2024

Optimizing Search Advertising Strategies: Integrating Reinforcement Learning with Generalized Second-Price Auctions for Enhanced Ad Ranking and Bidding

This paper explores the integration of strategic optimization methods in search advertising, focusing on ad ranking and bidding mechanisms within E-commerce platforms. By employing a combination of reinforcement learning and evolutionary strategies, we propose a dynamic model that adjusts to varying user interactions and optimizes the balance between advertiser cost, user relevance, and platform revenue. Our results suggest significant improvements in ad placement accuracy and cost efficiency, demonstrating the model's applicability in real-world scenarios.

cs.LG↗

arXiv2024

Online Merging Optimizers for Boosting Rewards and Mitigating Tax in Alignment

Effectively aligning Large Language Models (LLMs) with human-centric values while preventing the degradation of abilities acquired through Pre-training and Supervised Fine-tuning (SFT) poses a central challenge in Reinforcement Learning from Human Feedback (RLHF). In this paper, we first discover that interpolating RLHF and SFT model parameters can adjust the trade-off between human preference and basic capabilities, thereby reducing the alignment tax at the cost of alignment reward. Inspired by this, we propose integrating the RL policy and SFT models at each optimization step in RLHF to continuously regulate the training direction, introducing the Online Merging Optimizer. Specifically, we merge gradients with the parameter differences between SFT and pretrained models, effectively steering the gradient towards maximizing rewards in the direction of SFT optimization. We demonstrate that our optimizer works well with different LLM families, such as Qwen and LLaMA, across various model sizes ranging from 1.8B to 8B, various RLHF algorithms like DPO and KTO, and existing model merging methods. It significantly enhances alignment reward while mitigating alignment tax, achieving higher overall performance across 14 benchmarks.

cs.CL↗

arXiv2024

TD3 Based Collision Free Motion Planning for Robot Navigation

This paper addresses the challenge of collision-free motion planning in automated navigation within complex environments. Utilizing advancements in Deep Reinforcement Learning (DRL) and sensor technologies like LiDAR, we propose the TD3-DWA algorithm, an innovative fusion of the traditional Dynamic Window Approach (DWA) with the Twin Delayed Deep Deterministic Policy Gradient (TD3). This hybrid algorithm enhances the efficiency of robotic path planning by optimizing the sampling interval parameters of DWA to effectively navigate around both static and dynamic obstacles. The performance of the TD3-DWA algorithm is validated through various simulation experiments, demonstrating its potential to significantly improve the reliability and safety of autonomous navigation systems.

cs.RO↗

arXiv2024

Research on Image Recognition Technology Based on Multimodal Deep Learning

This project investigates the human multi-modal behavior identification algorithm utilizing deep neural networks. According to the characteristics of different modal information, different deep neural networks are used to adapt to different modal video information. Through the integration of various deep neural networks, the algorithm successfully identifies behaviors across multiple modalities. In this project, multiple cameras developed by Microsoft Kinect were used to collect corresponding bone point data based on acquiring conventional images. In this way, the motion features in the image can be extracted. Ultimately, the behavioral characteristics discerned through both approaches are synthesized to facilitate the precise identification and categorization of behaviors. The performance of the suggested algorithm was evaluated using the MSR3D data set. The findings from these experiments indicate that the accuracy in recognizing behaviors remains consistently high, suggesting that the algorithm is reliable in various scenarios. Additionally, the tests demonstrate that the algorithm substantially enhances the accuracy of detecting pedestrian behaviors in video footage.

cs.CV↗

arXiv2024

Adaptive speed planning for Unmanned Vehicle Based on Deep Reinforcement Learning

In order to solve the problem of frequent deceleration of unmanned vehicles when approaching obstacles, this article uses a Deep Q-Network (DQN) and its extension, the Double Deep Q-Network (DDQN), to develop a local navigation system that adapts to obstacles while maintaining optimal speed planning. By integrating improved reward functions and obstacle angle determination methods, the system demonstrates significant enhancements in maneuvering capabilities without frequent decelerations. Experiments conducted in simulated environments with varying obstacle densities confirm the effectiveness of the proposed method in achieving more stable and efficient path planning.

cs.RO↗

arXiv2024

Localization Through Particle Filter Powered Neural Network Estimated Monocular Camera Poses

The reduced cost and computational and calibration requirements of monocular cameras make them ideal positioning sensors for mobile robots, albeit at the expense of any meaningful depth measurement. Solutions proposed by some scholars to this localization problem involve fusing pose estimates from convolutional neural networks (CNNs) with pose estimates from geometric constraints on motion to generate accurate predictions of robot trajectories. However, the distribution of attitude estimation based on CNN is not uniform, resulting in certain translation problems in the prediction of robot trajectories. This paper proposes improving these CNN-based pose estimates by propagating a SE(3) uniform distribution driven by a particle filter. The particles utilize the same motion model used by the CNN, while updating their weights using CNN-based estimates. The results show that while the rotational component of pose estimation does not consistently improve relative to CNN-based estimation, the translational component is significantly more accurate. This factor combined with the superior smoothness of the filtered trajectories shows that the use of particle filters significantly improves the performance of CNN-based localization algorithms.

cs.RO↗

arXiv2024

Speculative Contrastive Decoding

Large language models~(LLMs) exhibit exceptional performance in language tasks, yet their auto-regressive inference is limited due to high computational requirements and is sub-optimal due to the exposure bias. Inspired by speculative decoding and contrastive decoding, we introduce Speculative Contrastive Decoding~(SCD), a straightforward yet powerful decoding approach that leverages predictions from smaller language models~(LMs) to achieve both decoding acceleration and quality improvement. Extensive evaluations and analyses on four diverse language tasks demonstrate the effectiveness of SCD, showing that decoding efficiency and quality can compatibly benefit from one smaller LM.

cs.CL↗

arXiv2024

Traffic Control via Connected and Automated Vehicles: An Open-Road Field Experiment with 100 CAVs

The CIRCLES project aims to reduce instabilities in traffic flow, which are naturally occurring phenomena due to human driving behavior. These "phantom jams" or "stop-and-go waves,"are a significant source of wasted energy. Toward this goal, the CIRCLES project designed a control system referred to as the MegaController by the CIRCLES team, that could be deployed in real traffic. Our field experiment leveraged a heterogeneous fleet of 100 longitudinally-controlled vehicles as Lagrangian traffic actuators, each of which ran a controller with the architecture described in this paper. The MegaController is a hierarchical control architecture, which consists of two main layers. The upper layer is called Speed Planner, and is a centralized optimal control algorithm. It assigns speed targets to the vehicles, conveyed through the LTE cellular network. The lower layer is a control layer, running on each vehicle. It performs local actuation by overriding the stock adaptive cruise controller, using the stock on-board sensors. The Speed Planner ingests live data feeds provided by third parties, as well as data from our own control vehicles, and uses both to perform the speed assignment. The architecture of the speed planner allows for modular use of standard control techniques, such as optimal control, model predictive control, kernel methods and others, including Deep RL, model predictive control and explicit controllers. Depending on the vehicle architecture, all onboard sensing data can be accessed by the local controllers, or only some. Control inputs vary across different automakers, with inputs ranging from torque or acceleration requests for some cars, and electronic selection of ACC set points in others. The proposed architecture allows for the combination of all possible settings proposed above. Most configurations were tested throughout the ramp up to the MegaVandertest.

eess.SY↗

arXiv2024

Sketch and Refine: Towards Fast and Accurate Lane Detection

Lane detection is to determine the precise location and shape of lanes on the road. Despite efforts made by current methods, it remains a challenging task due to the complexity of real-world scenarios. Existing approaches, whether proposal-based or keypoint-based, suffer from depicting lanes effectively and efficiently. Proposal-based methods detect lanes by distinguishing and regressing a collection of proposals in a streamlined top-down way, yet lack sufficient flexibility in lane representation. Keypoint-based methods, on the other hand, construct lanes flexibly from local descriptors, which typically entail complicated post-processing. In this paper, we present a "Sketch-and-Refine" paradigm that utilizes the merits of both keypoint-based and proposal-based methods. The motivation is that local directions of lanes are semantically simple and clear. At the "Sketch" stage, local directions of keypoints can be easily estimated by fast convolutional layers. Then we can build a set of lane proposals accordingly with moderate accuracy. At the "Refine" stage, we further optimize these proposals via a novel Lane Segment Association Module (LSAM), which allows adaptive lane segment adjustment. Last but not least, we propose multi-level feature integration to enrich lane feature representations more efficiently. Based on the proposed "Sketch and Refine" paradigm, we propose a fast yet effective lane detector dubbed "SRLane". Experiments show that our SRLane can run at a fast speed (i.e., 278 FPS) while yielding an F1 score of 78.9\%. The source code is available at: https://github.com/passerer/SRLane.

cs.CV↗

arXiv2024

AT2022cmc: a Tidal Disruption Event with Two-component Jet in a Bondi-profile Circumnuclear Medium

A supermassive black hole can launch a relativistic jet when it violently disrupts a star that passes too close. Such jetted tidal disruption events (TDEs) are rare and unique tools to investigate quiescent supermassive black holes, jet physics, and circumnuclear environment at high redshift. The newly discovered TDE AT2022cmc ($z\sim 1.193$) providing rich multi-band (X-ray, UV, optical, sub-millimeter, and radio) data, has been interpreted as the fourth on-axis jetted TDE. In this work, we constrain the circumnuclear medium (CNM) density profile with both closure relation (CR) test and detailed forward shock model fit with Markov chain Monte Carlo (MCMC) approach to the multi-band (optical, sub-millimeter, and radio) data of AT2022cmc.We find that the CNM density profile of AT2022cmc is $n\propto R^{-k}$ with $k \sim 1.68$, implying a Bondi accretion in history. Furthermore, our model fit result suggests a two-component jet in AT2022cmc, indicating a similar jet physics to well-studied jetted TDE Sw J1644+57.

astro-ph.HE↗