arXiv subjects

Di Wu

Publications and source records attributed to Di Wu.

At least 451 records · Page 25Linked to original sources

arXiv2024

Beyond Shared Vocabulary: Increasing Representational Word Similarities across Languages for Multilingual Machine Translation

Using a vocabulary that is shared across languages is common practice in Multilingual Neural Machine Translation (MNMT). In addition to its simple design, shared tokens play an important role in positive knowledge transfer, assuming that shared tokens refer to similar meanings across languages. However, when word overlap is small, especially due to different writing systems, transfer is inhibited. In this paper, we define word-level information transfer pathways via word equivalence classes and rely on graph networks to fuse word embeddings across languages. Our experiments demonstrate the advantages of our approach: 1) embeddings of words with similar meanings are better aligned across languages, 2) our method achieves consistent BLEU improvements of up to 2.3 points for high- and low-resource MNMT, and 3) less than 1.0\% additional trainable parameters are required with a limited increase in computational costs, while inference time remains identical to the baseline. We release the codebase to the community.

cs.CL↗

arXiv2024

Kernel-based multi-marker tests of association based on the accelerated failure time model

Kernel-based multi-marker tests for survival outcomes use primarily the Cox model to adjust for covariates. The proportional hazards assumption made by the Cox model could be unrealistic, especially in the long-term follow-up. We develop a suite of novel multi-marker survival tests for genetic association based on the accelerated failure time model, which is a popular alternative to the Cox model due to its direct physical interpretation. The tests are based on the asymptotic distributions of their test statistics and are thus computationally efficient. The association tests can account for the heterogeneity of genetic effects across sub-populations/individuals to increase the power. All the new tests can deal with competing risks and left truncation. Moreover, we develop small-sample corrections to the tests to improve their accuracy under small samples. Extensive numerical experiments show that the new tests perform very well in various scenarios. An application to a genetic dataset of Alzheimer's disease illustrates the tests' practical utility.

stat.ME↗

arXiv2024

Hallucination Detection and Hallucination Mitigation: An Investigation

Large language models (LLMs), including ChatGPT, Bard, and Llama, have achieved remarkable successes over the last two years in a range of different applications. In spite of these successes, there exist concerns that limit the wide application of LLMs. A key problem is the problem of hallucination. Hallucination refers to the fact that in addition to correct responses, LLMs can also generate seemingly correct but factually incorrect responses. This report aims to present a comprehensive review of the current literature on both hallucination detection and hallucination mitigation. We hope that this report can serve as a good reference for both engineers and researchers who are interested in LLMs and applying them to real world tasks.

cs.CL↗

arXiv2024

Masked Modeling for Self-supervised Representation Learning on Vision and Beyond

As the deep learning revolution marches on, self-supervised learning has garnered increasing attention in recent years thanks to its remarkable representation learning ability and the low dependence on labeled data. Among these varied self-supervised techniques, masked modeling has emerged as a distinctive approach that involves predicting parts of the original data that are proportionally masked during training. This paradigm enables deep models to learn robust representations and has demonstrated exceptional performance in the context of computer vision, natural language processing, and other modalities. In this survey, we present a comprehensive review of the masked modeling framework and its methodology. We elaborate on the details of techniques within masked modeling, including diverse masking strategies, recovering targets, network architectures, and more. Then, we systematically investigate its wide-ranging applications across domains. Furthermore, we also explore the commonalities and differences between masked modeling methods in different fields. Toward the end of this paper, we conclude by discussing the limitations of current techniques and point out several potential avenues for advancing masked modeling research. A paper list project with this survey is available at \url{https://github.com/Lupin1998/Awesome-MIM}.

cs.CV↗

arXiv2024

On the Selection of Intermediate Length Representative Periods for Capacity Expansion

As the decarbonization of power systems accelerates, there has been increasing interest in capacity expansion models for their role in guiding this transition. Representative period selection is an important component of capacity expansion modeling, enabling computational tractability of optimization while ensuring fidelity between the representative periods and the full year. However, little attention has been devoted to selecting representative periods longer than a single day. This prevents the capacity expansion model from directly simulating interday energy sharing, which is of key importance as energy generation becomes more variable and storage more important. To this end, we propose a novel method for selecting representative periods of any length. The method is validated using a capacity expansion model and production cost model based on California's decarbonization goals. We demonstrate that the representative period length has a substantial impact in the results of the capacity expansion investment plan.

eess.SY↗

arXiv2024

EcoFed: Efficient Communication for DNN Partitioning-based Federated Learning

Efficiently running federated learning (FL) on resource-constrained devices is challenging since they are required to train computationally intensive deep neural networks (DNN) independently. DNN partitioning-based FL (DPFL) has been proposed as one mechanism to accelerate training where the layers of a DNN (or computation) are offloaded from the device to the server. However, this creates significant communication overheads since the intermediate activation and gradient need to be transferred between the device and the server during training. While current research reduces the communication introduced by DNN partitioning using local loss-based methods, we demonstrate that these methods are ineffective in improving the overall efficiency (communication overhead and training speed) of a DPFL system. This is because they suffer from accuracy degradation and ignore the communication costs incurred when transferring the activation from the device to the server. This article proposes EcoFed - a communication efficient framework for DPFL systems. EcoFed eliminates the transmission of the gradient by developing pre-trained initialization of the DNN model on the device for the first time. This reduces the accuracy degradation seen in local loss-based methods. In addition, EcoFed proposes a novel replay buffer mechanism and implements a quantization-based compression technique to reduce the transmission of the activation. It is experimentally demonstrated that EcoFed can reduce the communication cost by up to 133x and accelerate training by up to 21x when compared to classic FL. Compared to vanilla DPFL, EcoFed achieves a 16x communication reduction and 2.86x training time speed-up. EcoFed is available from https://github.com/blessonvar/EcoFed.

cs.DC↗

arXiv2023

A Split-and-Privatize Framework for Large Language Model Fine-Tuning

Fine-tuning is a prominent technique to adapt a pre-trained language model to downstream scenarios. In parameter-efficient fine-tuning, only a small subset of modules are trained over the downstream datasets, while leaving the rest of the pre-trained model frozen to save computation resources. In recent years, a popular productization form arises as Model-as-a-Service (MaaS), in which vendors provide abundant pre-trained language models, server resources and core functions, and customers can fine-tune, deploy and invoke their customized model by accessing the one-stop MaaS with their own private dataset. In this paper, we identify the model and data privacy leakage risks in MaaS fine-tuning, and propose a Split-and-Privatize (SAP) framework, which manage to mitigate the privacy issues by adapting the existing split learning architecture. The proposed SAP framework is sufficiently investigated by experiments, and the results indicate that it can enhance the empirical privacy by 62% at the cost of 1% model performance degradation on the Stanford Sentiment Treebank dataset.

cs.CL↗

arXiv2023

Model-independent search for the quasinormal modes of gravitational wave echoes

Postmerger gravitational wave echoes provide a unique opportunity to probe the near-horizon structure of astrophysical black holes, which may be modified due to nonperturbative quantum gravity phenomena. However, since the waveform is subject to large theoretical uncertainties, it is necessary to develop search methods that are less reliant on specific models for detecting echoes from observational data. A promising strategy is to identify the characteristic quasinormal modes (QNMs) associated with echoes, {\it in frequency space}, which complements existing searches of quasiperiodic pulses in time. In this study, we build upon our previous work targeting these modes by incorporating relative phase information to optimize the Bayesian search algorithm. Using a new phase-marginalized likelihood, the performance can be significantly improved for well-resolved QNMs. This enables an efficient search for QNMs of various shapes, utilizing a simple search template that is independent of specific models. To demonstrate the robustness of the search algorithm, we construct four complementary benchmarks for the echo waveform that span a diverse range of different theoretical possibilities for the near-horizon structure. We then validate our Bayesian search algorithms by injecting the benchmark models into different realizations of Gaussian noise. Using two types of phase-marginalized likelihoods, we find that the search algorithm can efficiently detect the corresponding QNMs. Therefore, our search strategy provides a concrete Bayesian and model-independent approach to "quantum black hole seismology."

gr-qc↗

arXiv2023

Neural Network Approximation for Pessimistic Offline Reinforcement Learning

Deep reinforcement learning (RL) has shown remarkable success in specific offline decision-making scenarios, yet its theoretical guarantees are still under development. Existing works on offline RL theory primarily emphasize a few trivial settings, such as linear MDP or general function approximation with strong assumptions and independent data, which lack guidance for practical use. The coupling of deep learning and Bellman residuals makes this problem challenging, in addition to the difficulty of data dependence. In this paper, we establish a non-asymptotic estimation error of pessimistic offline RL using general neural network approximation with $\mathcal{C}$-mixing data regarding the structure of networks, the dimension of datasets, and the concentrability of data coverage, under mild assumptions. Our result shows that the estimation error consists of two parts: the first converges to zero at a desired rate on the sample size with partially controllable concentrability, and the second becomes negligible if the residual constraint is tight. This result demonstrates the explicit efficiency of deep adversarial offline RL frameworks. We utilize the empirical process tool for $\mathcal{C}$-mixing sequences and the neural network approximation theory for the Hölder class to achieve this. We also develop methods to bound the Bellman estimation error caused by function approximation with empirical Bellman constraint perturbations. Additionally, we present a result that lessens the curse of dimensionality using data with low intrinsic dimensionality and function classes with low complexity. Our estimation provides valuable insights into the development of deep offline RL and guidance for algorithm model design.

cs.LG↗

arXiv2023

An Iterative Bidirectional Gradient Boosting Approach for CVR Baseline Estimation

This paper presents a novel Iterative Bidirectional Gradient Boosting Model (IBi-GBM) for estimating the baseline of Conservation Voltage Reduction (CVR) programs. In contrast to many existing methods, we treat CVR baseline estimation as a missing data retrieval problem. The approach involves dividing the load and its corresponding temperature profiles into three periods: pre-CVR, CVR, and post-CVR. To restore the missing load profile during the CVR period, the method employs a three-step process. First, a forward-pass GBM is executed using data from the pre-CVR period as inputs. Subsequently, a backward-pass GBM is applied using data from the post-CVR period. The two restored load profiles are reconciled, considering pre-calculated weights derived from forecasting accuracy, and only the leftmost and rightmost points are retained. The newly restored points are then included as inputs for the subsequent iteration. This iterative procedure continues until the original load data in the CVR period is fully restored. We develop IBi-GBM using actual smart meter and Supervisory Control and Data Acquisition (SCADA) data. Our results demonstrate that IBi-GBM exhibits robust performance across various data resolutions and in different seasons and outperforms existing methods by achieving a 1-2% reduction in normalized Root Mean Square Error (nRMSE).

eess.SY↗

arXiv2023

Traffic Signal Control Using Lightweight Transformers: An Offline-to-Online RL Approach

Efficient traffic signal control is critical for reducing traffic congestion and improving overall transportation efficiency. The dynamic nature of traffic flow has prompted researchers to explore Reinforcement Learning (RL) for traffic signal control (TSC). Compared with traditional methods, RL-based solutions have shown preferable performance. However, the application of RL-based traffic signal controllers in the real world is limited by the low sample efficiency and high computational requirements of these solutions. In this work, we propose DTLight, a simple yet powerful lightweight Decision Transformer-based TSC method that can learn policy from easily accessible offline datasets. DTLight novelly leverages knowledge distillation to learn a lightweight controller from a well-trained larger teacher model to reduce implementation computation. Additionally, it integrates adapter modules to mitigate the expenses associated with fine-tuning, which makes DTLight practical for online adaptation with minimal computation and only a few fine-tuning steps during real deployment. Moreover, DTLight is further enhanced to be more applicable to real-world TSC problems. Extensive experiments on synthetic and real-world scenarios show that DTLight pre-trained purely on offline datasets can outperform state-of-the-art online RL-based methods in most scenarios. Experiment results also show that online fine-tuning further improves the performance of DTLight by up to 42.6% over the best online RL baseline methods. In this work, we also introduce Datasets specifically designed for TSC with offline RL (referred to as DTRL). Our datasets and code are publicly available.

cs.LG↗

arXiv2023

Anomaly Detection for Scalable Task Grouping in Reinforcement Learning-based RAN Optimization

The use of learning-based methods for optimizing cellular radio access networks (RAN) has received increasing attention in recent years. This coincides with a rapid increase in the number of cell sites worldwide, driven largely by dramatic growth in cellular network traffic. Training and maintaining learned models that work well across a large number of cell sites has thus become a pertinent problem. This paper proposes a scalable framework for constructing a reinforcement learning policy bank that can perform RAN optimization across a large number of cell sites with varying traffic patterns. Central to our framework is a novel application of anomaly detection techniques to assess the compatibility between sites (tasks) and the policy bank. This allows our framework to intelligently identify when a policy can be reused for a task, and when a new policy needs to be trained and added to the policy bank. Our results show that our approach to compatibility assessment leads to an efficient use of computational resources, by allowing us to construct a performant policy bank without exhaustively training on all tasks, which makes it applicable under real-world constraints.

cs.LG↗

arXiv2023

Attend Who is Weak: Enhancing Graph Condensation via Cross-Free Adversarial Training

In this paper, we study the \textit{graph condensation} problem by compressing the large, complex graph into a concise, synthetic representation that preserves the most essential and discriminative information of structure and features. We seminally propose the concept of Shock Absorber (a type of perturbation) that enhances the robustness and stability of the original graphs against changes in an adversarial training fashion. Concretely, (I) we forcibly match the gradients between pre-selected graph neural networks (GNNs) trained on a synthetic, simplified graph and the original training graph at regularly spaced intervals. (II) Before each update synthetic graph point, a Shock Absorber serves as a gradient attacker to maximize the distance between the synthetic dataset and the original graph by selectively perturbing the parts that are underrepresented or insufficiently informative. We iteratively repeat the above two processes (I and II) in an adversarial training fashion to maintain the highly-informative context without losing correlation with the original dataset. More importantly, our shock absorber and the synthesized graph parallelly share the backward process in a free training manner. Compared to the original adversarial training, it introduces almost no additional time overhead. We validate our framework across 8 datasets (3 graph and 5 node classification datasets) and achieve prominent results: for example, on Cora, Citeseer and Ogbn-Arxiv, we can gain nearly 1.13% to 5.03% improvements compare with SOTA models. Moreover, our algorithm adds only about 0.2% to 2.2% additional time overhead over Flicker, Citeseer and Ogbn-Arxiv. Compared to the general adversarial training, our approach improves time efficiency by nearly 4-fold.

cs.LG↗

arXiv2023

Online Two-stage Thermal History Prediction Method for Metal Additive Manufacturing of Thin Walls

This paper aims to propose an online two-stage thermal history prediction method, which could be integrated into a metal AM process for performance control. Based on the similarity of temperature curves (curve segments of a temperature profile of one point) between any two successive layers, the first stage of the proposed method designs a layer-to-layer prediction model to estimate the temperature curves of the yet-to-print layer from measured temperatures of certain points on the previously printed layer. With measured/predicted temperature profiles of several points on the same layer, the second stage proposes a reduced order model (ROM) (intra-layer prediction model) to decompose and construct the temperature profiles of all points on the same layer, which could be used to build the temperature field of the entire layer. The training of ROM is performed with an extreme learning machine (ELM) for computational efficiency. Fifteen wire arc AM experiments and nine simulations are designed for thin walls with a fixed length and unidirectional printing of each layer. The test results indicate that the proposed prediction method could construct the thermal history of a yet-to-print layer within 0.1 seconds on a low-cost desktop computer. Meanwhile, the method has acceptable generalization capability in most cases from lower layers to higher layers in the same simulation, as well as from one simulation to a new simulation on different AM process parameters. More importantly, after fine-tuning the proposed method with limited experimental data, the relative errors of all predicted temperature profiles on a new experiment are smaller than 0.09, which demonstrates the applicability and generalization of the proposed two-stage thermal history prediction method in online applications for metal AM.

cs.LG↗

arXiv2023

A Better Match for Drivers and Riders: Reinforcement Learning at Lyft

To better match drivers to riders in our ridesharing application, we revised Lyft's core matching algorithm. We use a novel online reinforcement learning approach that estimates the future earnings of drivers in real time and use this information to find more efficient matches. This change was the first documented implementation of a ridesharing matching algorithm that can learn and improve in real time. We evaluated the new approach during weeks of switchback experimentation in most Lyft markets, and estimated how it benefited drivers, riders, and the platform. In particular, it enabled our drivers to serve millions of additional riders each year, leading to more than $30 million per year in incremental revenue. Lyft rolled out the algorithm globally in 2021.

cs.LG↗

arXiv2023

CeBed: A Benchmark for Deep Data-Driven OFDM Channel Estimation

Deep learning has been extensively used in wireless communication problems, including channel estimation. Although several data-driven approaches exist, a fair and realistic comparison between them is difficult due to inconsistencies in the experimental conditions and the lack of a standardized experimental design. In addition, the performance of data-driven approaches is often compared based on empirical analysis. The lack of reproducibility and availability of standardized evaluation tools (e.g., datasets, codebases) hinder the development and progress of data-driven methods for channel estimation and wireless communication in general. In this work, we introduce an initiative to build benchmarks that unify several data-driven OFDM channel estimation approaches. Specifically, we present CeBed (a testbed for channel estimation) including different datasets covering various systems models and propagation conditions along with the implementation of ten deep and traditional baselines. This benchmark considers different practical aspects such as the robustness of the data-driven models, the number and the arrangement of pilots, and the number of receive antennas. This work offers a comprehensive and unified framework to help researchers evaluate and design data-driven channel estimation algorithms.

cs.AI↗

arXiv2023

Insight-HXMT on-orbit thermal control status and thermal deformation impact analysis

Purpose: The Hard X-ray Modulation Telescope is China's first X-ray astronomy satellite launched on June 15th, 2017, dubbed Insight-HXMT. Active and passive thermal control measures are employed to keep devices at suitable temperatures. In this paper, we analyzed the on-orbit thermal monitoring data of the first 5 years and investigated the effect of thermal deformation on the point spread function (PSF) of the telescopes. Methods: We examined the data of the on-orbit temperatures measured using 157 thermistors placed on the collimators, detectors and their support structures and compared the results with the thermal control requirements. The thermal deformation was evaluated by the relative orientation of the two star sensors installed on the main support structure. its effect was estimated with evolution of the PSF obtained with calibration scanning observations of the Crab nebula. Conclusion: The on-orbit temperatures met the thermal control requirements thus far, and the effect of thermal deformation on the PSF was negligible after the on-orbit pointing calibration.

astro-ph.IM↗

arXiv2023

New constraints on primordial non-Gaussianity from missing two-loop contributions of scalar induced gravitational waves

We analyze the energy density spectrum of \acp{SIGW} using the NANOGrav 15-year data set, thereby constraining the primordial non-Gaussian parameter $f_{\mathrm{NL}}$. For the first time, we calculate the seventeen missing two-loop diagrams proportional to $f_{\mathrm{NL}}A_ζ^3$ that correspond to the two-point correlation function $\langle h^{λ,(3)}_{\mathbf{k}} h^{λ',(2)}_{\mathbf{k}'} \rangle$ for local-type primordial non-Gaussianity. The total energy density spectrum of \acp{SIGW} can be significantly suppressed by these two-loop diagrams. If \acp{SIGW} dominate the \acp{SGWB} observed in \ac{PTA} experiments, the parameter interval $f_{\mathrm{NL}}\in [-5,-1]$ is notably excluded based on NANOGrav 15-year data set. After taking into account abundance of \acp{PBH} and the convergence of the cosmological perturbation expansion, we find that the only possible parameter range for $f_{\mathrm{NL}}$ might be $-1\le f_{\mathrm{NL}}< 0$.

astro-ph.CO↗