In this work, we study how the performance of a given direction changes with its sampling ratio in Multilingual Neural Machine Translation (MNMT). By training over 200 multilingual models with various model sizes, data sizes, and language directions, we find it interesting that the performance of certain translation direction does not always improve with the increase of its weight in the multi-task optimization objective. Accordingly, scalarization method leads to a multitask trade-off front that deviates from the traditional Pareto front when there exists data imbalance in the training corpus, which poses a great challenge to improve the overall performance of all directions. Based on our observations, we propose the Double Power Law to predict the unique performance trade-off front in MNMT, which is robust across various languages, data adequacy, and the number of tasks. Finally, we formulate the sample ratio selection problem in MNMT as an optimization problem based on the Double Power Law. In our experiments, it achieves better performance than temperature searching and gradient manipulation methods with only 1/5 to 1/2 of the total training budget. We release the code at //github.com/pkunlp-icler/ParetoMNMT for reproduction.
相關內容
The purpose of this article is to investigate the viability of Multi-Carrier Modulation (MCM) systems based on the Fast Walsh Hadamard Transform (FWHT). In addition, a nonlinear Joint Low-Complexity Optimized Zero Forcing Successive Interference Cancellation (JLCOZF-SIC) equalizer is proposed. To that end, general equations for the number of flops of the proposed equalizer and various other equalizers are given. This article discusses the use of Banded Matrix Approximation (BMA) as a technique for reducing complexity. The proposed equalizer uses BMA to accomplish both equalization and co-Carrier Frequency Offset (co-CFO) corrections. In addition, three cases involving the proposed equalizer were investigated. In the first case, diagonal compensation is used. In the second case, BMA compensation is used. In the third case, complete matrix compensation is used. In the presence of frequency offset, noise, and frequency-selective Rayleigh fading environments, analysis and simulation results show that the OFDM-FWHT system with the proposed equalizer outperforms the conventional OFDM system with various linear and nonlinear equalizers.
With the rapid progress in virtual reality (VR) technology, the scope of VR applications has greatly expanded across various domains. However, the superiority of VR training over traditional methods and its impact on learning efficacy are still uncertain. To investigate whether VR training is more effective than traditional methods, we designed virtual training systems for mechanical assembly on both VR and desktop platforms, subsequently conducting pre-test and post-test experiments. A cohort of 53 students, all enrolled in engineering drawing course without prior knowledge distinctions, was randomly divided into three groups: physical training, desktop virtual training, and immersive VR training. Our investigation utilized analysis of covariance (ANCOVA) to examine the differences in post-test scores among the three groups while controlling for pre-test scores. The group that received VR training showed the highest scores on the post-test. Another facet of our study delved into the presence of the virtual system. We developed a specialized scale to assess this aspect for our research objectives. Our findings indicate that VR training can enhance the sense of presence, particularly in terms of sensory factors and realism factors. Moreover, correlation analysis uncovers connections between the various dimensions of presence. This study confirms that using VR training can improve learning efficacy and the presence in the context of mechanical assembly, surpassing traditional training methods. Furthermore, it provides empirical evidence supporting the integration of VR technology in higher education and engineering training. This serves as a reference for the practical application of VR technology in different fields.
The Vision Transformer has emerged as a powerful tool for image classification tasks, surpassing the performance of convolutional neural networks (CNNs). Recently, many researchers have attempted to understand the robustness of Transformers against adversarial attacks. However, previous researches have focused solely on perturbations in the spatial domain. This paper proposes an additional perspective that explores the adversarial robustness of Transformers against frequency-selective perturbations in the spectral domain. To facilitate comparison between these two domains, an attack framework is formulated as a flexible tool for implementing attacks on images in the spatial and spectral domains. The experiments reveal that Transformers rely more on phase and low frequency information, which can render them more vulnerable to frequency-selective attacks than CNNs. This work offers new insights into the properties and adversarial robustness of Transformers.
This paper studies the problem of forecasting general stochastic processes using a path-dependent extension of the Neural Jump ODE (NJ-ODE) framework \citep{herrera2021neural}. While NJ-ODE was the first framework to establish convergence guarantees for the prediction of irregularly observed time series, these results were limited to data stemming from It\^o-diffusions with complete observations, in particular Markov processes, where all coordinates are observed simultaneously. In this work, we generalise these results to generic, possibly non-Markovian or discontinuous, stochastic processes with incomplete observations, by utilising the reconstruction properties of the signature transform. These theoretical results are supported by empirical studies, where it is shown that the path-dependent NJ-ODE outperforms the original NJ-ODE framework in the case of non-Markovian data. Moreover, we show that PD-NJ-ODE can be applied successfully to classical stochastic filtering problems and to limit order book (LOB) data.
In this work, we present a new visual prompting method called 3DAxiesPrompts (3DAP) to unleash the capabilities of GPT-4V in performing 3D spatial tasks. Our investigation reveals that while GPT-4V exhibits proficiency in discerning the position and interrelations of 2D entities through current visual prompting techniques, its abilities in handling 3D spatial tasks have yet to be explored. In our approach, we create a 3D coordinate system tailored to 3D imagery, complete with annotated scale information. By presenting images infused with the 3DAP visual prompt as inputs, we empower GPT-4V to ascertain the spatial positioning information of the given 3D target image with a high degree of precision. Through experiments, We identified three tasks that could be stably completed using the 3DAP method, namely, 2D to 3D Point Reconstruction, 2D to 3D point matching, and 3D Object Detection. We perform experiments on our proposed dataset 3DAP-Data, the results from these experiments validate the efficacy of 3DAP-enhanced GPT-4V inputs, marking a significant stride in 3D spatial task execution.
Visual Place Recognition (VPR) is a critical task for performing global re-localization in visual perception systems. It requires the ability to accurately recognize a previously visited location under variations such as illumination, occlusion, appearance and viewpoint. In the case of robotic systems and augmented reality, the target devices for deployment are battery powered edge devices. Therefore whilst the accuracy of VPR methods is important so too is memory consumption and latency. Recently new works have focused on the recall@1 metric as a performance measure with limited focus on resource utilization. This has resulted in methods that use deep learning models too large to deploy on low powered edge devices. We hypothesize that these large models are highly over-parameterized and can be optimized to satisfy the constraints of a low powered embedded system whilst maintaining high recall performance. Our work studies the impact of compact convolutional network architecture design in combination with full-precision and mixed-precision post-training quantization on VPR performance. Importantly we not only measure performance via the recall@1 score but also measure memory consumption and latency. We characterize the design implications on memory, latency and recall scores and provide a number of design recommendations for VPR systems under these resource limitations.
Graph Neural Networks (GNNs) have paved its way for being a cornerstone in graph related learning tasks. From a theoretical perspective, the expressive power of GNNs is primarily characterised according to their ability to distinguish non-isomorphic graphs. It is a well-known fact that most of the conventional GNNs are upper-bounded by Weisfeiler-Lehman graph isomorphism test (1-WL). In this work, we study the expressive power of graph neural networks through the lens of graph partitioning. This follows from our observation that permutation invariant graph partitioning enables a powerful way of exploring structural interactions among vertex sets and subgraphs, and can help uplifting the expressive power of GNNs efficiently. Based on this, we first establish a theoretical connection between graph partitioning and graph isomorphism. Then we introduce a novel GNN architecture, namely Graph Partitioning Neural Networks (GPNNs). We theoretically analyse how a graph partitioning scheme and different kinds of structural interactions relate to the k-WL hierarchy. Empirically, we demonstrate its superior performance over existing GNN models in a variety of graph benchmark tasks.
We present a novel method, the Chain of Empathy (CoE) prompting, that utilizes insights from psychotherapy to induce Large Language Models (LLMs) to reason about human emotional states. This method is inspired by various psychotherapy approaches including Cognitive Behavioral Therapy (CBT), Dialectical Behavior Therapy (DBT), Person Centered Therapy (PCT), and Reality Therapy (RT), each leading to different patterns of interpreting clients' mental states. LLMs without reasoning generated predominantly exploratory responses. However, when LLMs used CoE reasoning, we found a more comprehensive range of empathetic responses aligned with the different reasoning patterns of each psychotherapy model. The CBT based CoE resulted in the most balanced generation of empathetic responses. The findings underscore the importance of understanding the emotional context and how it affects human and AI communication. Our research contributes to understanding how psychotherapeutic models can be incorporated into LLMs, facilitating the development of context-specific, safer, and empathetic AI.
The dominating NLP paradigm of training a strong neural predictor to perform one task on a specific dataset has led to state-of-the-art performance in a variety of applications (eg. sentiment classification, span-prediction based question answering or machine translation). However, it builds upon the assumption that the data distribution is stationary, ie. that the data is sampled from a fixed distribution both at training and test time. This way of training is inconsistent with how we as humans are able to learn from and operate within a constantly changing stream of information. Moreover, it is ill-adapted to real-world use cases where the data distribution is expected to shift over the course of a model's lifetime. The first goal of this thesis is to characterize the different forms this shift can take in the context of natural language processing, and propose benchmarks and evaluation metrics to measure its effect on current deep learning architectures. We then proceed to take steps to mitigate the effect of distributional shift on NLP models. To this end, we develop methods based on parametric reformulations of the distributionally robust optimization framework. Empirically, we demonstrate that these approaches yield more robust models as demonstrated on a selection of realistic problems. In the third and final part of this thesis, we explore ways of efficiently adapting existing models to new domains or tasks. Our contribution to this topic takes inspiration from information geometry to derive a new gradient update rule which alleviate catastrophic forgetting issues during adaptation.
Machine learning techniques have deeply rooted in our everyday life. However, since it is knowledge- and labor-intensive to pursue good learning performance, human experts are heavily involved in every aspect of machine learning. In order to make machine learning techniques easier to apply and reduce the demand for experienced human experts, automated machine learning (AutoML) has emerged as a hot topic with both industrial and academic interest. In this paper, we provide an up to date survey on AutoML. First, we introduce and define the AutoML problem, with inspiration from both realms of automation and machine learning. Then, we propose a general AutoML framework that not only covers most existing approaches to date but also can guide the design for new methods. Subsequently, we categorize and review the existing works from two aspects, i.e., the problem setup and the employed techniques. Finally, we provide a detailed analysis of AutoML approaches and explain the reasons underneath their successful applications. We hope this survey can serve as not only an insightful guideline for AutoML beginners but also an inspiration for future research.
小貼士
登錄享
注冊地址： 北京市海淀區羊坊店路18號2幢3層301-191