arXiv subjects

Xiuyuan Lu

Publications and source records attributed to Xiuyuan Lu.

At least 19 recordsLinked to original sources

arXiv2026

Efficient Exploration at Scale

We develop an online learning algorithm that dramatically improves the data efficiency of reinforcement learning from human feedback (RLHF). Our algorithm incrementally updates reward and language models as choice data is received. The reward model is fit to the choice data, while the language model is updated by a variation of reinforce, with reinforcement signals provided by the reward model. Several features enable the efficiency gains: a small affirmative nudge added to each reinforcement signal, an epistemic neural network that models reward uncertainty, and information-directed exploration. With Gemma large language models (LLMs), our algorithm matches the performance of offline RLHF trained on 200K labels using fewer than 20K labels, representing more than a 10x gain in data efficiency. Extrapolating from our results, we expect our algorithm trained on 1M labels to match offline RLHF trained on 1B labels. This represents a 1,000x gain. To our knowledge, these are the first results to demonstrate that such large improvements are possible.

cs.LG↗

arXiv2025

ESVO2: Direct Visual-Inertial Odometry with Stereo Event Cameras

Event-based visual odometry is a specific branch of visual Simultaneous Localization and Mapping (SLAM) techniques, which aims at solving tracking and mapping subproblems (typically in parallel), by exploiting the special working principles of neuromorphic (i.e., event-based) cameras. Due to the motion-dependent nature of event data, explicit data association (i.e., feature matching) under large-baseline view-point changes is difficult to establish, making direct methods a more rational choice. However, state-of-the-art direct methods are limited by the high computational complexity of the mapping sub-problem and the degeneracy of camera pose tracking in certain degrees of freedom (DoF) in rotation. In this paper, we tackle these issues by building an event-based stereo visual-inertial odometry system on top of a direct pipeline. Specifically, to speed up the mapping operation, we propose an efficient strategy for sampling contour points according to the local dynamics of events. The mapping performance is also improved in terms of structure completeness and local smoothness by merging the temporal stereo and static stereo results. To circumvent the degeneracy of camera pose tracking in recovering the pitch and yaw components of general 6-DoF motion, we introduce IMU measurements as motion priors via pre-integration. To this end, a compact back-end is proposed for continuously updating the IMU bias and predicting the linear velocity, enabling an accurate motion prediction for camera pose tracking. The resulting system scales well with modern high-resolution event cameras and leads to better global positioning accuracy in large-scale outdoor environments. Extensive evaluations on five publicly available datasets featuring different resolutions and scenarios justify the superior performance of the proposed system against five state-of-the-art methods.

cs.CV↗

arXiv2025

Temporal and Rotational Calibration for Event-Centric Multi-Sensor Systems

Event cameras generate asynchronous signals in response to pixel-level brightness changes, offering a sensing paradigm with theoretically microsecond-scale latency that can significantly enhance the performance of multi-sensor systems. Extrinsic calibration is a critical prerequisite for effective sensor fusion; however, the configuration that involves event cameras remains an understudied topic. In this paper, we propose a motion-based temporal and rotational calibration framework tailored for event-centric multi-sensor systems, eliminating the need for dedicated calibration targets. Our method uses as input the rotational motion estimates obtained from event cameras and other heterogeneous sensors, respectively. Different from conventional approaches that rely on event-to-frame conversion, our method efficiently estimates angular velocity from normal flow observations, which are derived from the spatio-temporal profile of event data. The overall calibration pipeline adopts a two-step approach: it first initializes the temporal offset and rotational extrinsics by exploiting kinematic correlations in the spirit of Canonical Correlation Analysis (CCA), and then refines both temporal and rotational parameters through a joint non-linear optimization using a continuous-time parametrization in SO(3). Extensive evaluations on both publicly available and self-collected datasets validate that the proposed method achieves calibration accuracy comparable to target-based methods, while exhibiting superior stability over purely CCA-based methods, and highlighting its precision, robustness and flexibility. To facilitate future research, our implementation will be made open-source. Code: https://github.com/NAIL-HNU/EvMultiCalib.

cs.RO↗

arXiv2024

High-Performance Green and Blue Light-Emitting Diodes Enabled by CdZnSe/ZnS Core/Shell Colloidal Quantum Wells

The unique anisotropic properties of colloidal quantum wells (CQWs) make them highly promising as components in nanocrystal-based devices. However, the limited performance of green and blue light-emitting diodes (LEDs) based on CQWs has impeded their practical applications. In this study, we tailored alloy CdZnSe core CQWs with precise compositions via direct cation exchange (CE) from CdSe CQWs with specific size, shape, and crystal structure and utilized hot-injection shell (HIS) growth to synthesize CdZnSe/ZnS core/shell CQWs exhibiting exceptional optoelectronic characteristics. This approach enabled us to successfully fabricate green and blue LEDs manifesting superior performance compared to previously reported solution-processed CQW-LEDs. Our devices demonstrated a remarkable peak external quantum efficiency (20.4% for green and 10.6% for blue), accompanied by a maximum brightness 347,683 cd m-2 for green and 38,063 cd m-2 for blue. The high-performance represents a significant advancement for nanocrystal-based light-emitting diodes (Nc-LEDs) incorporating anisotropic nanocrystals. This work provides a comprehensive synthesis strategy for enhancing the efficiency of Nc-LEDs utilizing anisotropic nanocrystals.

physics.optics↗

arXiv2024

Water-induced high-performance quantum-dot light-emitting diodes

Solution-processed light-emitting diodes (LEDs) are appealing for their potential in the low-cost fabrication of large-area devices. However, the limited performance of solution-processed blue LEDs, particularly their short operation lifetime, is hindering their practical use in display technologies. Here, we demonstrate that trace water in device, previously considered detrimental to most solution-processed LEDs, dramatically enhances the performance of quantum-dot LEDs (QLEDs). This breakthrough stems from our comprehensive mechanism investigations into the positive ageing phenomenon, a long-standing puzzle in the QLED field. Our findings reveal that water passivation on the surface of electron-transport layers, which are composed of zinc-oxide-based nanoparticles, improves charge transport and enhances exciton radiative recombination during device operation. Combined with the advanced top-emitting architecture, our blue QLEDs achieve a high current efficiency of 35.5 cd A-1, a blue index (colour coordinate corrected current efficiency) of over 470 cd A-1 CIEy-1, and unprecedented stability, with an extrapolated T95 lifetime (at an initial brightness of 1,000 cd m-2) of 287 hours. Our work may inspire further exploration into surface passivation of nanocrystalline functional layers, critical for the advancement of emerging solution-processed optoelectronic and electronic devices.

physics.optics↗

arXiv2024

Event-based Visual Inertial Velometer

Neuromorphic event-based cameras are bio-inspired visual sensors with asynchronous pixels and extremely high temporal resolution. Such favorable properties make them an excellent choice for solving state estimation tasks under aggressive ego motion. However, failures of camera pose tracking are frequently witnessed in state-of-the-art event-based visual odometry systems when the local map cannot be updated in time. One of the biggest roadblocks for this specific field is the absence of efficient and robust methods for data association without imposing any assumption on the environment. This problem seems, however, unlikely to be addressed as in standard vision due to the motion-dependent observability of event data. Therefore, we propose a mapping-free design for event-based visual-inertial state estimation in this paper. Instead of estimating the position of the event camera, we find that recovering the instantaneous linear velocity is more consistent with the differential working principle of event cameras. The proposed event-based visual-inertial velometer leverages a continuous-time formulation that incrementally fuses the heterogeneous measurements from a stereo event camera and an inertial measurement unit. Experiments on the synthetic dataset demonstrate that the proposed method can recover instantaneous linear velocity in metric scale with low latency.

cs.RO↗

arXiv2024

RLHF and IIA: Perverse Incentives

Existing algorithms for reinforcement learning from human feedback (RLHF) can incentivize responses at odds with preferences because they are based on models that assume independence of irrelevant alternatives (IIA). The perverse incentives induced by IIA hinder innovations on query formats and learning algorithms.

cs.LG↗

arXiv2023

Epistemic Neural Networks

Intelligence relies on an agent's knowledge of what it does not know. This capability can be assessed based on the quality of joint predictions of labels across multiple inputs. In principle, ensemble-based approaches produce effective joint predictions, but the computational costs of training large ensembles can become prohibitive. We introduce the epinet: an architecture that can supplement any conventional neural network, including large pretrained models, and can be trained with modest incremental computation to estimate uncertainty. With an epinet, conventional neural networks outperform very large ensembles, consisting of hundreds or more particles, with orders of magnitude less computation. The epinet does not fit the traditional framework of Bayesian neural networks. To accommodate development of approaches beyond BNNs, such as the epinet, we introduce the epistemic neural network (ENN) as an interface for models that produce joint predictions.

cs.LG↗

arXiv2023

Reinforcement Learning, Bit by Bit

Reinforcement learning agents have demonstrated remarkable achievements in simulated environments. Data efficiency poses an impediment to carrying this success over to real environments. The design of data-efficient agents calls for a deeper understanding of information acquisition and representation. We discuss concepts and regret analysis that together offer principled guidance. This line of thinking sheds light on questions of what information to seek, how to seek that information, and what information to retain. To illustrate concepts, we design simple agents that build on them and present computational results that highlight data efficiency.

cs.LG↗

arXiv2023

Ensemble Sampling

Thompson sampling has emerged as an effective heuristic for a broad range of online decision problems. In its basic form, the algorithm requires computing and sampling from a posterior distribution over models, which is tractable only for simple special cases. This paper develops ensemble sampling, which aims to approximate Thompson sampling while maintaining tractability even in the face of complex models such as neural networks. Ensemble sampling dramatically expands on the range of applications for which Thompson sampling is viable. We establish a theoretical basis that supports the approach and present computational results that offer further insight.

stat.ML↗

arXiv2023

An Analysis of Ensemble Sampling

Ensemble sampling serves as a practical approximation to Thompson sampling when maintaining an exact posterior distribution over model parameters is computationally intractable. In this paper, we establish a regret bound that ensures desirable behavior when ensemble sampling is applied to the linear bandit problem. This represents the first rigorous regret analysis of ensemble sampling and is made possible by leveraging information-theoretic concepts and novel analytic techniques that may prove useful beyond the scope of this paper.

cs.LG↗

arXiv2023

Approximate Thompson Sampling via Epistemic Neural Networks

Thompson sampling (TS) is a popular heuristic for action selection, but it requires sampling from a posterior distribution. Unfortunately, this can become computationally intractable in complex environments, such as those modeled using neural networks. Approximate posterior samples can produce effective actions, but only if they reasonably approximate joint predictive distributions of outputs across inputs. Notably, accuracy of marginal predictive distributions does not suffice. Epistemic neural networks (ENNs) are designed to produce accurate joint predictive distributions. We compare a range of ENNs through computational experiments that assess their performance in approximating TS across bandit and reinforcement learning environments. The results indicate that ENNs serve this purpose well and illustrate how the quality of joint predictive distributions drives performance. Further, we demonstrate that the \textit{epinet} -- a small additive network that estimates uncertainty -- matches the performance of large ensembles at orders of magnitude lower computational cost. This enables effective application of TS with computation that scales gracefully to complex environments.

cs.LG↗

arXiv2022

The Neural Testbed: Evaluating Joint Predictions

Predictive distributions quantify uncertainties ignored by point estimates. This paper introduces The Neural Testbed: an open-source benchmark for controlled and principled evaluation of agents that generate such predictions. Crucially, the testbed assesses agents not only on the quality of their marginal predictions per input, but also on their joint predictions across many inputs. We evaluate a range of agents using a simple neural network data generating process. Our results indicate that some popular Bayesian deep learning agents do not fare well with joint predictions, even when they can produce accurate marginal predictions. We also show that the quality of joint predictions drives performance in downstream decision tasks. We find these results are robust across choice a wide range of generative models, and highlight the practical importance of joint predictions to the community.

cs.LG↗

arXiv2022

Robustness of Epinets against Distributional Shifts

Recent work introduced the epinet as a new approach to uncertainty modeling in deep learning. An epinet is a small neural network added to traditional neural networks, which, together, can produce predictive distributions. In particular, using an epinet can greatly improve the quality of joint predictions across multiple inputs, a measure of how well a neural network knows what it does not know. In this paper, we examine whether epinets can offer similar advantages under distributional shifts. We find that, across ImageNet-A/O/C, epinets generally improve robustness metrics. Moreover, these improvements are more significant than those afforded by even very large ensembles at orders of magnitude lower computational costs. However, these improvements are relatively small compared to the outstanding issues in distributionally-robust deep learning. Epinets may be a useful tool in the toolbox, but they are far from the complete solution.

cs.LG↗

arXiv2022

Ensembles for Uncertainty Estimation: Benefits of Prior Functions and Bootstrapping

In machine learning, an agent needs to estimate uncertainty to efficiently explore and adapt and to make effective decisions. A common approach to uncertainty estimation maintains an ensemble of models. In recent years, several approaches have been proposed for training ensembles, and conflicting views prevail with regards to the importance of various ingredients of these approaches. In this paper, we aim to address the benefits of two ingredients -- prior functions and bootstrapping -- which have come into question. We show that prior functions can significantly improve an ensemble agent's joint predictions across inputs and that bootstrapping affords additional benefits if the signal-to-noise ratio varies across inputs. Our claims are justified by both theoretical and experimental results.

cs.LG↗

arXiv2022

From Predictions to Decisions: The Importance of Joint Predictive Distributions

A fundamental challenge for any intelligent system is prediction: given some inputs, can you predict corresponding outcomes? Most work on supervised learning has focused on producing accurate marginal predictions for each input. However, we show that for a broad class of decision problems, accurate joint predictions are required to deliver good performance. In particular, we establish several results pertaining to combinatorial decision problems, sequential predictions, and multi-armed bandits to elucidate the essential role of joint predictive distributions. Our treatment of multi-armed bandits introduces an approximate Thompson sampling algorithm and analytic techniques that lead to a new kind of regret bound.

cs.LG↗

arXiv2022

Evaluating High-Order Predictive Distributions in Deep Learning

Most work on supervised learning research has focused on marginal predictions. In decision problems, joint predictive distributions are essential for good performance. Previous work has developed methods for assessing low-order predictive distributions with inputs sampled i.i.d. from the testing distribution. With low-dimensional inputs, these methods distinguish agents that effectively estimate uncertainty from those that do not. We establish that the predictive distribution order required for such differentiation increases greatly with input dimension, rendering these methods impractical. To accommodate high-dimensional inputs, we introduce \textit{dyadic sampling}, which focuses on predictive distributions associated with random \textit{pairs} of inputs. We demonstrate that this approach efficiently distinguishes agents in high-dimensional examples involving simple logistic regression as well as complex synthetic and empirical data.

stat.ML↗

arXiv2021

Event-based Motion Segmentation by Cascaded Two-Level Multi-Model Fitting

Among prerequisites for a synthetic agent to interact with dynamic scenes, the ability to identify independently moving objects is specifically important. From an application perspective, nevertheless, standard cameras may deteriorate remarkably under aggressive motion and challenging illumination conditions. In contrast, event-based cameras, as a category of novel biologically inspired sensors, deliver advantages to deal with these challenges. Its rapid response and asynchronous nature enables it to capture visual stimuli at exactly the same rate of the scene dynamics. In this paper, we present a cascaded two-level multi-model fitting method for identifying independently moving objects (i.e., the motion segmentation problem) with a monocular event camera. The first level leverages tracking of event features and solves the feature clustering problem under a progressive multi-model fitting scheme. Initialized with the resulting motion model instances, the second level further addresses the event clustering problem using a spatio-temporal graph-cut method. This combination leads to efficient and accurate event-wise motion segmentation that cannot be achieved by any of them alone. Experiments demonstrate the effectiveness and versatility of our method in real-world scenes with different motion patterns and an unknown number of independently moving objects.

cs.CV↗