arXiv subjects

Chao Yu

Publications and source records attributed to Chao Yu.

At least 127 records · Page 7Linked to original sources

arXiv2024

Revisiting Discrete Soft Actor-Critic

We study the adaption of Soft Actor-Critic (SAC), which is considered as a state-of-the-art reinforcement learning (RL) algorithm, from continuous action space to discrete action space. We revisit vanilla discrete SAC and provide an in-depth understanding of its Q value underestimation and performance instability issues when applied to discrete settings. We thereby propose Stable Discrete SAC (SDSAC), an algorithm that leverages entropy-penalty and double average Q-learning with Q-clip to address these issues. Extensive experiments on typical benchmarks with discrete action space, including Atari games and a large-scale MOBA game, show the efficacy of our proposed method. Our code is at: https://github.com/coldsummerday/SD-SAC.git.

cs.LG↗

arXiv2024

Hunyuan-Large: An Open-Source MoE Model with 52 Billion Activated Parameters by Tencent

In this paper, we introduce Hunyuan-Large, which is currently the largest open-source Transformer-based mixture of experts model, with a total of 389 billion parameters and 52 billion activation parameters, capable of handling up to 256K tokens. We conduct a thorough evaluation of Hunyuan-Large's superior performance across various benchmarks including language understanding and generation, logical reasoning, mathematical problem-solving, coding, long-context, and aggregated tasks, where it outperforms LLama3.1-70B and exhibits comparable performance when compared to the significantly larger LLama3.1-405B model. Key practice of Hunyuan-Large include large-scale synthetic data that is orders larger than in previous literature, a mixed expert routing strategy, a key-value cache compression technique, and an expert-specific learning rate strategy. Additionally, we also investigate the scaling laws and learning rate schedule of mixture of experts models, providing valuable insights and guidances for future model development and optimization. The code and checkpoints of Hunyuan-Large are released to facilitate future innovations and applications. Codes: https://github.com/Tencent/Hunyuan-Large Models: https://huggingface.co/tencent/Tencent-Hunyuan-Large

cs.CL↗

arXiv2024

SleepNetZero: Zero-Burden Zero-Shot Reliable Sleep Staging With Neural Networks Based on Ballistocardiograms

Sleep monitoring plays a crucial role in maintaining good health, with sleep staging serving as an essential metric in the monitoring process. Traditional methods, utilizing medical sensors like EEG and ECG, can be effective but often present challenges such as unnatural user experience, complex deployment, and high costs. Ballistocardiography~(BCG), a type of piezoelectric sensor signal, offers a non-invasive, user-friendly, and easily deployable alternative for long-term home monitoring. However, reliable BCG-based sleep staging is challenging due to the limited sleep monitoring data available for BCG. A restricted training dataset prevents the model from generalization across populations. Additionally, transferring to BCG faces difficulty ensuring model robustness when migrating from other data sources. To address these issues, we introduce SleepNetZero, a zero-shot learning based approach for sleep staging. To tackle the generalization challenge, we propose a series of BCG feature extraction methods that align BCG components with corresponding respiratory, cardiac, and movement channels in PSG. This allows models to be trained on large-scale PSG datasets that are diverse in population. For the migration challenge, we employ data augmentation techniques, significantly enhancing generalizability. We conducted extensive training and testing on large datasets~(12393 records from 9637 different subjects), achieving an accuracy of 0.803 and a Cohen's Kappa of 0.718. ZeroSleepNet was also deployed in real prototype~(monitoring pads) and tested in actual hospital settings~(265 users), demonstrating an accuracy of 0.697 and a Cohen's Kappa of 0.589. To the best of our knowledge, this work represents the first known reliable BCG-based sleep staging effort and marks a significant step towards in-home health monitoring.

eess.SP↗

arXiv2024

Diverse Policies Recovering via Pointwise Mutual Information Weighted Imitation Learning

Recovering a spectrum of diverse policies from a set of expert trajectories is an important research topic in imitation learning. After determining a latent style for a trajectory, previous diverse policies recovering methods usually employ a vanilla behavioral cloning learning objective conditioned on the latent style, treating each state-action pair in the trajectory with equal importance. Based on an observation that in many scenarios, behavioral styles are often highly relevant with only a subset of state-action pairs, this paper presents a new principled method in diverse polices recovery. In particular, after inferring or assigning a latent style for a trajectory, we enhance the vanilla behavioral cloning by incorporating a weighting mechanism based on pointwise mutual information. This additional weighting reflects the significance of each state-action pair's contribution to learning the style, thus allowing our method to focus on state-action pairs most representative of that style. We provide theoretical justifications for our new objective, and extensive empirical evaluations confirm the effectiveness of our method in recovering diverse policies from expert data.

cs.LG↗

arXiv2024

Reward-Robust RLHF in LLMs

As Large Language Models (LLMs) continue to progress toward more advanced forms of intelligence, Reinforcement Learning from Human Feedback (RLHF) is increasingly seen as a key pathway toward achieving Artificial General Intelligence (AGI). However, the reliance on reward-model-based (RM-based) alignment methods introduces significant challenges due to the inherent instability and imperfections of Reward Models (RMs), which can lead to critical issues such as reward hacking and misalignment with human intentions. In this paper, we introduce a reward-robust RLHF framework aimed at addressing these fundamental challenges, paving the way for more reliable and resilient learning in LLMs. Our approach introduces a novel optimization objective that carefully balances performance and robustness by incorporating Bayesian Reward Model Ensembles (BRME) to model the uncertainty set of reward functions. This allows the framework to integrate both nominal performance and minimum reward signals, ensuring more stable learning even with imperfect RMs. Empirical results demonstrate that our framework consistently outperforms baselines across diverse benchmarks, showing improved accuracy and long-term stability. We also provide a theoretical analysis, demonstrating that reward-robust RLHF approaches the stability of constant reward settings, which proves to be acceptable even in a stochastic-case analysis. Together, these contributions highlight the framework potential to enhance both the performance and stability of LLM alignment.

cs.LG↗

arXiv2024

Is DPO Superior to PPO for LLM Alignment? A Comprehensive Study

Reinforcement Learning from Human Feedback (RLHF) is currently the most widely used method to align large language models (LLMs) with human preferences. Existing RLHF methods can be roughly categorized as either reward-based or reward-free. Novel applications such as ChatGPT and Claude leverage reward-based methods that first learn a reward model and apply actor-critic algorithms, such as Proximal Policy Optimization (PPO). However, in academic benchmarks, state-of-the-art results are often achieved via reward-free methods, such as Direct Preference Optimization (DPO). Is DPO truly superior to PPO? Why does PPO perform poorly on these benchmarks? In this paper, we first conduct both theoretical and empirical studies on the algorithmic properties of DPO and show that DPO may have fundamental limitations. Moreover, we also comprehensively examine PPO and reveal the key factors for the best performances of PPO in fine-tuning LLMs. Finally, we benchmark DPO and PPO across a collection of RLHF testbeds, ranging from dialogue to code generation. Experiment results demonstrate that PPO is able to surpass other alignment methods in all cases and achieve state-of-the-art results in challenging code competitions. Our code is publicly available at https://github.com/openpsi-project/ReaLHF.

cs.CL↗

arXiv2024

Trajectory Tracking for MmWave Communication Systems via Cooperative Passive Sensing

In this paper, a cooperative passive sensing framework for millimeter wave (mmWave) communication systems is proposed and demonstrated in a scenario with one mobile signal blocker. Specifically, in the uplink communication with at least two transmitters, a cooperative detection method is proposed for the receiver to track the blocker's trajectory, localize the transmitters and detect the potential link blockage jointly. To facilitate detection, the receiver collects the signal of each transmitter along a line-of-sight (LoS) path and a non-line-of-sight (NLoS) path separately via two narrow-beam phased arrays. The latter path should scatter at the mobile blocker, and hence it can be identified by the Doppler frequency. Comparing the received signals of both paths, the Doppler frequency and angle-of-arrival (AoA) of the NLoS path can be estimated. To resolve the blocker's trajectory and the transmitters' locations, the receiver should continuously track the mobile blocker to accumulate sufficient numbers of the Doppler frequency and AoA versus time observations. Finally, a gradient-descent-based algorithm is proposed for joint detection. With the reconstructed trajectory, the potential link blockage can be predicted. It is demonstrated that the system can achieve decimeter-level localization and trajectory estimation, and predict the blockage time with an error less than 0.1 s.

eess.SP↗

arXiv2024

An Offline Adaptation Framework for Constrained Multi-Objective Reinforcement Learning

In recent years, significant progress has been made in multi-objective reinforcement learning (RL) research, which aims to balance multiple objectives by incorporating preferences for each objective. In most existing studies, specific preferences must be provided during deployment to indicate the desired policies explicitly. However, designing these preferences depends heavily on human prior knowledge, which is typically obtained through extensive observation of high-performing demonstrations with expected behaviors. In this work, we propose a simple yet effective offline adaptation framework for multi-objective RL problems without assuming handcrafted target preferences, but only given several demonstrations to implicitly indicate the preferences of expected policies. Additionally, we demonstrate that our framework can naturally be extended to meet constraints on safety-critical objectives by utilizing safe demonstrations, even when the safety thresholds are unknown. Empirical results on offline multi-objective and safe tasks demonstrate the capability of our framework to infer policies that align with real preferences while meeting the constraints implied by the provided demonstrations.

cs.LG↗

arXiv2024

Recent advances in InGaAs/InP single-photon detectors

Single-photon detectors (SPDs) are widely used in applications requiring extremely weak light detection. In the near-infrared region, SPDs based on InGaAs/InP single-photon avalanche diodes (SPADs) are the primary candidates for practical applications because of their small size, low cost and ease of operation. Driven by the escalating demands for quantum communication and lidar, the performance of InGaAs/InP SPDs has been continuously enhanced. This paper provides a comprehensive review of advances in InGaAs/InP SPDs over the past 10 years, including the investigation into SPAD structures and mechanisms, as well as emerging readout techniques for both gated and free-running mode SPDs. In addition, future prospects are also summarised.

physics.ins-det↗

arXiv2024

LAGOON: Language-Guided Motion Control

We aim to control a robot to physically behave in the real world following any high-level language command like "cartwheel" or "kick". Although human motion datasets exist, this task remains particularly challenging since generative models can produce physically unrealistic motions, which will be more severe for robots due to different body structures and physical properties. Deploying such a motion to a physical robot can cause even greater difficulties due to the sim2real gap. We develop LAnguage-Guided mOtion cONtrol (LAGOON), a multi-phase reinforcement learning (RL) method to generate physically realistic robot motions under language commands. LAGOON first leverages a pretrained model to generate a human motion from a language command. Then an RL phase trains a control policy in simulation to mimic the generated human motion. Finally, with domain randomization, our learned policy can be deployed to a quadrupedal robot, leading to a quadrupedal robot that can take diverse behaviors in the real world under natural language commands

cs.RO↗

arXiv2024

A Dual Curriculum Learning Framework for Multi-UAV Pursuit-Evasion in Diverse Environments

This paper addresses multi-UAV pursuit-evasion, where a group of drones cooperates to capture a fast evader in a confined environment with obstacles. Existing heuristic algorithms, which simplify the pursuit-evasion problem, often lack expressive coordination strategies and struggle to capture the evader in extreme scenarios, such as when the evader moves at high speeds. In contrast, reinforcement learning (RL) has been applied to this problem and has the potential to obtain highly cooperative capture strategies. However, RL-based methods face challenges in training for complex 3-dimensional scenarios with diverse task settings due to the vast exploration space. The dynamics constraints of drones further restrict the ability of reinforcement learning to acquire high-performance capture strategies. In this work, we introduce a dual curriculum learning framework, named DualCL, which addresses multi-UAV pursuit-evasion in diverse environments and demonstrates zero-shot transfer ability to unseen scenarios. DualCL comprises two main components: the Intrinsic Parameter Curriculum Proposer, which progressively suggests intrinsic parameters from easy to hard to improve the capture capability of drones, and the External Environment Generator, tasked with exploring unresolved scenarios and generating appropriate training distributions of external environment parameters. The simulation experimental results show that DualCL significantly outperforms baseline methods, achieving over 90% capture rate and reducing the capture timestep by at least 27.5% in the training scenarios. Additionally, it exhibits the best zero-shot generalization ability in unseen environments. Moreover, we demonstrate the transferability of our pursuit strategy from simulation to real-world environments. Further details can be found on the project website at https://sites.google.com/view/dualcl.

cs.LG↗

arXiv2024

CASTER: A Computer-Vision-Assisted Wireless Channel Simulator for Gesture Recognition

In this paper, a computer-vision-assisted simulation method is proposed to address the issue of training dataset acquisition for wireless hand gesture recognition. In the existing literature, in order to classify gestures via the wireless channel estimation, massive training samples should be measured in a consistent environment, consuming significant efforts. In the proposed CASTER simulator, however, the training dataset can be simulated via existing videos. Particularly, a gesture is represented by a sequence of snapshots, and the channel impulse response of each snapshot is calculated via tracing the rays scattered off a primitive-based hand model. Moreover, CASTER simulator relies on the existing videos to extract the motion data of gestures. Thus, the massive measurements of wireless channel can be eliminated. The experiments demonstrate a 90.8% average classification accuracy of simulation-to-reality inference.

eess.SP↗

arXiv2024

Off-Policy Primal-Dual Safe Reinforcement Learning

Primal-dual safe RL methods commonly perform iterations between the primal update of the policy and the dual update of the Lagrange Multiplier. Such a training paradigm is highly susceptible to the error in cumulative cost estimation since this estimation serves as the key bond connecting the primal and dual update processes. We show that this problem causes significant underestimation of cost when using off-policy methods, leading to the failure to satisfy the safety constraint. To address this issue, we propose conservative policy optimization, which learns a policy in a constraint-satisfying area by considering the uncertainty in cost estimation. This improves constraint satisfaction but also potentially hinders reward maximization. We then introduce local policy convexification to help eliminate such suboptimality by gradually reducing the estimation uncertainty. We provide theoretical interpretations of the joint coupling effect of these two ingredients and further verify them by extensive experiments. Results on benchmark tasks show that our method not only achieves an asymptotic performance comparable to state-of-the-art on-policy methods while using much fewer samples, but also significantly reduces constraint violation during training. Our code is available at https://github.com/ZifanWu/CAL.

cs.LG↗

arXiv2024

Miniaturized time-correlated single-photon counting module for time-of-flight non-line-of-sight imaging applications

Single-photon time-of-flight (TOF) non-line-of-sight (NLOS) imaging enables the high-resolution reconstruction of objects outside the field of view. The compactness of TOF NLOS imaging systems, entailing the miniaturization of key components within such systems is crucial for practical applications. Here, we present a miniaturized four-channel time-correlated single-photon counting module dedicated to TOF NLOS imaging applications. The module achieves excellent performance with a 10 ps bin size and 27.4 ps minimum root-mean-square time resolution. We present the results of TOF NLOS imaging experiment using an InGaAs/InP single-photon detector and the time-correlated single-photon counting module, and show that a 6.3 cm lateral resolution and 2.3 cm depth resolution can be achieved under the conditions of 5 m imaging distance and 1 ms pixel dwell time.

physics.ins-det↗

arXiv2024

Multi-Agent Reinforcement Learning with a Hierarchy of Reward Machines

In this paper, we study the cooperative Multi-Agent Reinforcement Learning (MARL) problems using Reward Machines (RMs) to specify the reward functions such that the prior knowledge of high-level events in a task can be leveraged to facilitate the learning efficiency. Unlike the existing work that RMs have been incorporated into MARL for task decomposition and policy learning in relatively simple domains or with an assumption of independencies among the agents, we present Multi-Agent Reinforcement Learning with a Hierarchy of RMs (MAHRM) that is capable of dealing with more complex scenarios when the events among agents can occur concurrently and the agents are highly interdependent. MAHRM exploits the relationship of high-level events to decompose a task into a hierarchy of simpler subtasks that are assigned to a small group of agents, so as to reduce the overall computational complexity. Experimental results in three cooperative MARL domains show that MAHRM outperforms other MARL methods using the same prior knowledge of high-level events.

cs.AI↗

arXiv2024

Safe Offline Reinforcement Learning with Real-Time Budget Constraints

Aiming at promoting the safe real-world deployment of Reinforcement Learning (RL), research on safe RL has made significant progress in recent years. However, most existing works in the literature still focus on the online setting where risky violations of the safety budget are likely to be incurred during training. Besides, in many real-world applications, the learned policy is required to respond to dynamically determined safety budgets (i.e., constraint threshold) in real time. In this paper, we target at the above real-time budget constraint problem under the offline setting, and propose Trajectory-based REal-time Budget Inference (TREBI) as a novel solution that models this problem from the perspective of trajectory distribution and solves it through diffusion model planning. Theoretically, we prove an error bound of the estimation on the episodic reward and cost under the offline setting and thus provide a performance guarantee for TREBI. Empirical results on a wide range of simulation tasks and a real-world large-scale advertising application demonstrate the capability of TREBI in solving real-time budget constraint problems under offline settings.

cs.LG↗

arXiv2024

The prominent and heterogeneous gender disparities in scientific novelty: evidence from biomedical doctoral theses

Scientific novelty is the essential driving force for research breakthroughs and innovation. However, little is known about how early-career scientists pursue novel research paths, and the gender disparities in this process. To address this research gap, this study investigates a comprehensive dataset of 279,424 doctoral theses in biomedical sciences authored by US Ph.D. graduates. Spanning from 1980 to 2016, the data originates from the ProQuest Dissertations & Theses Database. This study aims to shed light on Ph.D. students' pursuit of scientific novelty in their doctoral theses and assess gender-related differences in this process. Using a combinatorial approach and a pre-trained Bio-BERT model, we quantify the scientific novelty of doctoral theses based on bio-entities. Applying fractional logistic and quantile regression models, this study reveals a decreasing trend in scientific novelty over time and heterogeneous gender disparities in doctoral theses. Specifically, female students consistently exhibited lower scientific novelty levels than their male peers. When supervised by female advisors, students' theses are found to be less novel than those under male advisors. The significant interaction effect of female students and female advisors suggests that female advisors may amplify the gender disparity in scientific novelty. Moreover, heterogeneous gender disparities in scientific novelty are identified, with non-top-tier universities displaying more pronounced disparities, while the differences at higher percentile ranges were comparatively more minor. These findings indicate a potential underrepresentation of female scientists pursuing novel research during the early stages of their careers. Notably, the outcomes of this study hold significant policy implications for advancing the careers of female scientists.

physics.soc-ph↗

arXiv2024

On the Theory of Solid-State Harmonic Generation Governed by Crystal Symmetry

The solid-state harmonic generation (SSHG) derives from photocurrent coherence. The crystal symmetry, including point-group symmetry and time-reversal symmetry, constrains the amplitude and phase of the photocurrent, thus manipulates the coherent processes in SSHG. We revisit the expression of photocurrent under the electric dipole approximation and give an unambiguous picture of non-equilibrium dynamics of photocarriers on laser-dressed effective bands. In addition to the dynamical phase, we reveal the indispensable roles of the phase difference of transition dipole moments and the phase induced by shift vector in the photocurrent coherence. Microscopic mechanism of the selection rule, orientation dependence, polarization characteristics, time-frequency analysis and ellipticity dependence of harmonics governed by symmetries is uniformly clarified in our theoretical framework. This work integrates non-equilibrium electronic dynamics of condensed matter in strong laser fields, and paves a way to explore more nonlinear optical phenomena governed by crystal symmetry.

physics.optics↗