arXiv subjects

Le He

Publications and source records attributed to Le He.

9 recordsLinked to original sources

arXiv2026

Generative Actor-Critic with Soft Bridge Policies

Expressive generative policies such as diffusion and flow models are appealing for MaxEnt online reinforcement learning because of their ability to model multimodal and highly non-Gaussian action distributions. However, training effective soft generative policies faces two obstacles that often arise together. First, marginal action densities are often unavailable, so existing methods typically rely on entropy bounds, heuristic proxies or approximations. Second, iterative shared-parameter samplers raise inference cost and require backpropagation through time over repeated network evaluations, increasing memory cost and destabilizing policy optimization. These obstacles motivate us to seek a generative policy that exposes a tractable MaxEnt objective while requiring only a single sampled actor forward pass for action generation. To this end, we propose soft generative actor-critic (SoftGAC), whose actor defines a stochastic bridge from a fixed base latent to a terminal action latent in pre-tanh space. This structured bridge allows us to lift the MaxEnt objective as an analytically tractable path-wise relative-entropy objective against a high-entropy reference process. In practical finite-step implementation, this relative entropy reduces exactly to sampled transition control energy and thus provides principled soft regularization. Moreover, we keep the single-pass actor lightweight by using small step-specific bridge transitions, each evaluated only once per sampled action, while maintaining a parameter budget comparable to strong actor baselines. Extensive experiments on challenging continuous-control benchmarks show that SoftGAC attains higher or competitive returns than strong generative policy baselines, including diffusion and flow-matching policies, while staying in the low-latency regime of one-pass actors and showing considerable improvements in the compute-return tradeoff.

cs.LG↗

arXiv2025

Asynchronous Risk-Aware Multi-Agent Packet Routing for Ultra-Dense LEO Satellite Networks

The rise of ultra-dense LEO constellations creates a complex and asynchronous network environment, driven by their massive scale, dynamic topologies, and significant delays. This unique complexity demands an adaptive packet routing algorithm that is asynchronous, risk-aware, and capable of balancing diverse and often conflicting QoS objectives in a decentralized manner. However, existing methods fail to address this need, as they typically rely on impractical synchronous decision-making and/or risk-oblivious approaches. To tackle this gap, we introduce PRIMAL, an event-driven multi-agent routing framework designed specifically to allow each satellite to act independently on its own event-driven timeline, while managing the risk of worst-case performance degradation via a principled primal-dual approach. This is achieved by enabling agents to learn the full cost distribution of the targeted QoS objectives and constrain tail-end risks. Extensive simulations on a LEO constellation with 1584 satellites validate its superiority in effectively optimizing latency and balancing load. Compared to a recent risk-oblivious baseline, it reduces queuing delay by over 70%, and achieves a nearly 12 ms end-to-end delay reduction in loaded scenarios. This is accomplished by resolving the core conflict between naive shortest-path finding and congestion avoidance, highlighting such autonomous risk-awareness as a key to robust routing.

cs.NI↗

arXiv2025

SCA-LLM: Spectral-Attentive LLM-Based Wireless World Modeling for Agentic Communications

Future AI-native wireless networks are moving from reactive optimization to agentic decision-making that can sense, predict, and plan under fast-varying channels. This calls for wireless world models that can predict and roll out channel dynamics, for which multi-step channel state information (CSI) prediction offers a practical short-horizon look-ahead. Recent advances in foundation sequence models further motivate large language models (LLMs) as general-purpose dynamics learners when suitably adapted to non-text time-series signals. However, bridging CSI to LLMs is non-trivial because an effective adapter must expose informative spectral and temporal evolution patterns, while prior designs provide limited inductive bias to capture such channel structures. To this end, we propose SCA-LLM, a spectral-attentive LLM-based wireless world modeling framework that bridges CSI to LLMs via a spectral-channel attention (SCA) adapter. Specifically, the SCA adapter performs multi-spectral representation learning to extract informative channel features and align CSI with the LLM's sequence modeling capability, enabling parameter-efficient adaptation while keeping the LLM backbone largely frozen. Extensive simulations show that SCA-LLM achieves state-of-the-art prediction performance and strong zero-shot generalization, yielding up to -2.4 dB normalized mean squared error (NMSE) advantage over the previous LLM based method. Our ablation studies further confirm the effectiveness of the proposed SCA adapter in mitigating domain mismatch.

cs.IT↗

arXiv2024

Will you donate money to a chatbot? The effect of chatbot anthropomorphic features and persuasion strategies on willingness to donate

This work investigates the causal mechanism behind the effect of chatbot personification and persuasion strategies on users' perceptions and donation likelihood. In a 2 (personified vs. non-personified chatbot) x 2 (emotional vs. logical persuasion strategy) between-subjects experiment (N=76), participants engaged with a chatbot that represented a non-profit charitable organization. The results suggest that interaction with a personified chatbot evokes perceived anthropomorphism; however, it does not elicit greater willingness to donate. In fact, we found that commonly used anthropomorphic features, like name and narrative, led to negative attitudes toward an AI agent in the donation context. Our results showcase a preference for non-personified chatbots paired with logical persuasion appeal, emphasizing the significance of consistency in chatbot interaction, mirroring human-human engagement. We discuss the importance of moving from exploring the common scenario of a chatbot with machine identity vs. a chatbot with human identity in light of the recent regulations of AI systems.

cs.HC↗

arXiv2021

Learning based signal detection for MIMO systems with unknown noise statistics

This paper aims to devise a generalized maximum likelihood (ML) estimator to robustly detect signals with unknown noise statistics in multiple-input multiple-output (MIMO) systems. In practice, there is little or even no statistical knowledge on the system noise, which in many cases is non-Gaussian, impulsive and not analyzable. Existing detection methods have mainly focused on specific noise models, which are not robust enough with unknown noise statistics. To tackle this issue, we propose a novel ML detection framework to effectively recover the desired signal. Our framework is a fully probabilistic one that can efficiently approximate the unknown noise distribution through a normalizing flow. Importantly, this framework is driven by an unsupervised learning approach, where only the noise samples are required. To reduce the computational complexity, we further present a low-complexity version of the framework, by utilizing an initial estimation to reduce the search space. Simulation results show that our framework outperforms other existing algorithms in terms of bit error rate (BER) in non-analytical noise environments, while it can reach the ML performance bound in analytical noise environments. The code of this paper is available at https://github.com/skypitcher/manfe.

cs.LG↗

arXiv2021

Towards Optimally Efficient Search with Deep Learning for Large-Scale MIMO Systems

This paper investigates the optimal signal detection problem with a particular interest in large-scale multiple-input multiple-output (MIMO) systems. The problem is NP-hard and can be solved optimally by searching the shortest path on the decision tree. Unfortunately, the existing optimal search algorithms often involve prohibitively high complexities, which indicates that they are infeasible in large-scale MIMO systems. To address this issue, we propose a general heuristic search algorithm, namely, hyper-accelerated tree search (HATS) algorithm. The proposed algorithm employs a deep neural network (DNN) to estimate the optimal heuristic, and then use the estimated heuristic to speed up the underlying memory-bounded search algorithm. This idea is inspired by the fact that the underlying heuristic search algorithm reaches the optimal efficiency with the optimal heuristic function. Simulation results show that the proposed algorithm reaches almost the optimal bit error rate (BER) performance in large-scale systems, while the memory size can be bounded. In the meanwhile, it visits nearly the fewest tree nodes. This indicates that the proposed algorithm reaches almost the optimal efficiency in practical scenarios, and thereby it is applicable for large-scale systems. Besides, the code for this paper is available at \url{https://github.com/skypitcher/hats}.

cs.LG↗

arXiv2019

$L^{p}$ regularity of weighted Bergman projection on Fock-Bargmann-Hartogs domain

The Fock-Bargmann-Hartogs domain $D_{n, m}(\mu)$ is defined by $$ D_{n, m}(\mu):=\{(z, w)\in\mathbb{C}^{n}\times\mathbb{C}^m:\Vert w \Vert^2 0.$ The Fock-Bargmann-Hartogs domain $D_{n, m}(\mu)$ is an unbounded strongly pseudoconvex domain with smooth real-analytic boundary. In this paper, we first compute the weighted Bergman kernel of $D_{n, m}(\mu)$ with respect to the weight $(-\rho)^{\alpha}$, where $\rho(z,w):=\|w\|^2-e^{-\mu \|z\|^2}$ is a defining function for $D_{n, m}(\mu)$ and $\alpha>-1$. Then, for $p\in [1,\infty),$ we show that the corresponding weighted Bergman projection $P_{D_{n, m}(\mu), (-\rho)^{\alpha}}$ is unbounded on $L^p(D_{n, m}(\mu), (-\rho)^{\alpha})$, except for the trivial case $p=2$. In particular, this paper gives an example of an unbounded strongly pseudoconvex domain whose ordinary Bergman projection is $L^p$ irregular when $p\in [1,\infty)\setminus\{2\}$. This result turns out to be completely different from the well-known positive $L^p$ regularity result on bounded strongly pseudoconvex domain.

math.CV↗

arXiv2018

Ambient Electro-Synthesis of Ammonia - Electrode Porosity and Composition Engineering

Ammonia, key precursor for fertilizer production, convenient hydrogen carrier and emerging clean fuel, plays a pivotal role in sustaining life on earth. Currently, the main route for NH3 synthesis is via the heterogeneous catalytic Haber-Bosch process (N2+3H2 - 2NH3), which proceeds under extreme conditions of temperature and pressure with a very large carbon footprint. Herein we report that a pristine nitrogen-doped nanoporous graphitic carbon membrane (NCM) can electrochemically convert N2 into NH3 in an aqueous acidic solution under ambient conditions. The Faradaic efficiency and rate of production of NH3 on the NCM electrode reach 5.2% and 0.08 g m-2 h-1, respectively. After functionalization of the NCM with Au nanoparticles (Au NPs) these performance metrics are dramatically enhanced to 22% and 0.36 g m-2 h-1, respectively. These efficiencies and rates for the production of NH3 at room temperature and atmospheric pressure are unprecedented. As this system offers the potential to be scaled to industrial proportions there is a high likelihood it might displace the century old Haber-Bosch process.

physics.chem-ph↗

arXiv2016

Light Drag Effect of Vacuum Tube Versus Light Propagation in Stationary Vacuum Tube with Moving Source and Receiver

We presented a new way to examine the principle of relativity of Special Relativity. According to the principle of relativity, the light dragging by moving media and the light propagation in stationary media with moving source and receiver should be two totally equivalent phenomena. We select a vacuum tube with two glass rods at two ends as the optical media. The length of the middle vacuum cell is L and the thicknesses of the glass rods with refractive index n are D1 and D2. The light drag effect of the moving vacuum tube with speed v is a first-order effect, delta t = 2(n-1)(D1+D2)v/c^2, which is independent of L because vacuum does not perform a drag effect. Predicted by the principle of relativity, the change of the light propagation time interval with stationary vacuum tube and moving source and receiver must be the same, i.e., delta tao = delta t = 2(n-1)(D1+D2)v/c^2. However all analyses have shown that the change of the propagation time interval delta tao is caused by the motion of the receiver during the light propagation in the vacuum tube. Thus, the contribution of the glass rods in delta tao is 2n(D1+D2)v/c^2, not 2(n-1)(D1+D2)v/c^2 in delta t. Importantly, the contribution of the vacuum cell in delta tao is 2Lv/c^2, not zero in delta t. Our analyses are solid in optics. The genuine tests of the prediction of the principle of relativity can be conducted by the experiments with two atomic clocks, or the experiments with fiber Sagnac interferometers.

physics.optics↗