PyTorchKR Papers

최신 논문

총 9,380편
태그·기간 등 상세 필터

카테고리 탐색

로그인 후 이 태그 구독

arXiv:2207.01209

Sustainable AI Processing at the Edge

Edge computing is a popular target for accelerating machine learning algorithms supporting mobile devices without requiring the communication latencies to handle them in the cloud. Edge deployments of machine learning primarily consider traditional concerns such as SWaP constraints (Size, Weight, and Power) for their installations. However, such metrics are not entirely sufficient to consider environmental impacts from computing given the significant contributions from embodied energy and carbon. In this paper we explore the tradeoffs of convolutional neural network acceleration engines for both inference and on-line training. In particular, we explore the use of processing-in-memory (PIM) approaches, mobile GPU accelerators, and recently released FPGAs, and compare them with novel Racetrack memory PIM. Replacing PIM-enabled DDR3 with Racetrack memory PIM can recover its embodied energy as quickly as 1 year. For high activity ratios, mobile GPUs can be more sustainable but have higher embodied energy to overcome compared to PIM-enabled Racetrack memory.

arXiv:2207.01059

Identifying the Context Shift between Test Benchmarks and Production Data

Machine learning models are often brittle on production data despite achieving high accuracy on benchmark datasets. Benchmark datasets have traditionally served dual purposes: first, benchmarks offer a standard on which machine learning researchers can compare different methods, and second, benchmarks provide a model, albeit imperfect, of the real world. The incompleteness of test benchmarks (and the data upon which models are trained) hinder robustness in machine learning, enable shortcut learning, and leave models systematically prone to err on out-of-distribution and adversarially perturbed data. The mismatch between a single static benchmark dataset and a production dataset has traditionally been described as a dataset shift. In an effort to clarify how to address the mismatch between test benchmarks and production data, we introduce context shift to describe semantically meaningful changes in the underlying data generation process. Moreover, we identify three methods for addressing context shift that would otherwise lead to model prediction errors: first, we describe how human intuition and expert knowledge can identify semantically meaningful features upon which models systematically fail, second, we detail how dynamic benchmarking - with its focus on capturing the data generation process - can promote generalizability through corroboration, and third, we highlight that clarifying a model's limitations can reduce unexpected errors. Robust machine learning is focused on model performance beyond benchmarks, and as such, we consider three model organism domains - facial expression recognition, deepfake detection, and medical diagnosis - to highlight how implicit assumptions in benchmark tasks lead to errors in practice. By paying close attention to the role of context, researchers can design more comprehensive benchmarks, reduce context shift errors, and increase generalizability.

arXiv:2207.00556

Learning to correct spectral methods for simulating turbulent flows

Despite their ubiquity throughout science and engineering, only a handful of partial differential equations (PDEs) have analytical, or closed-form solutions. This motivates a vast amount of classical work on numerical simulation of PDEs and more recently, a whirlwind of research into data-driven techniques leveraging machine learning (ML). A recent line of work indicates that a hybrid of classical numerical techniques and machine learning can offer significant improvements over either approach alone. In this work, we show that the choice of the numerical scheme is crucial when incorporating physics-based priors. We build upon Fourier-based spectral methods, which are known to be more efficient than other numerical schemes for simulating PDEs with smooth and periodic solutions. Specifically, we develop ML-augmented spectral solvers for three common PDEs of fluid dynamics. Our models are more accurate (2-4x) than standard spectral solvers at the same resolution but have longer overall runtimes (~2x), due to the additional runtime cost of the neural network component. We also demonstrate a handful of key design principles for combining machine learning and numerical methods for solving PDEs.

arXiv:2207.00551공식 코드 · Papers with Code

Evaluating the Explainers: Black-Box Explainable Machine Learning for Student Success Prediction in MOOCs

Neural networks are ubiquitous in applied machine learning for education. Their pervasive success in predictive performance comes alongside a severe weakness, the lack of explainability of their decisions, especially relevant in human-centric fields. We implement five state-of-the-art methodologies for explaining black-box machine learning models (LIME, PermutationSHAP, KernelSHAP, DiCE, CEM) and examine the strengths of each approach on the downstream task of student performance prediction for five massive open online courses. Our experiments demonstrate that the families of explainers do not agree with each other on feature importance for the same Bidirectional LSTM models with the same representative set of students. We use Principal Component Analysis, Jensen-Shannon distance, and Spearman's rank-order correlation to quantitatively cross-examine explanations across methods and courses. Furthermore, we validate explainer performance across curriculum-based prerequisite relationships. Our results come to the concerning conclusion that the choice of explainer is an important decision and is in fact paramount to the interpretation of the predictive results, even more so than the course the model is trained on. Source code and models are released at http://github.com/epfl-ml4ed/evaluating-explainers.

arXiv:2207.00529공식 코드 · Papers with Code

Deep Learning and Symbolic Regression for Discovering Parametric Equations

Symbolic regression is a machine learning technique that can learn the governing formulas of data and thus has the potential to transform scientific discovery. However, symbolic regression is still limited in the complexity and dimensionality of the systems that it can analyze. Deep learning on the other hand has transformed machine learning in its ability to analyze extremely complex and high-dimensional datasets. We propose a neural network architecture to extend symbolic regression to parametric systems where some coefficient may vary but the structure of the underlying governing equation remains constant. We demonstrate our method on various analytic expressions, ODEs, and PDEs with varying coefficients and show that it extrapolates well outside of the training domain. The neural network-based architecture can also integrate with other deep learning architectures so that it can analyze high-dimensional data while being trained end-to-end. To this end we integrate our architecture with convolutional neural networks to analyze 1D images of varying spring systems.

arXiv:2207.00489

Panning for gold: Lessons learned from the platform-agnostic automated detection of political content in textual data

The growing availability of data about online information behaviour enables new possibilities for political communication research. However, the volume and variety of these data makes them difficult to analyse and prompts the need for developing automated content approaches relying on a broad range of natural language processing techniques (e.g. machine learning- or neural network-based ones). In this paper, we discuss how these techniques can be used to detect political content across different platforms. Using three validation datasets, which include a variety of political and non-political textual documents from online platforms, we systematically compare the performance of three groups of detection techniques relying on dictionaries, supervised machine learning, or neural networks. We also examine the impact of different modes of data preprocessing (e.g. stemming and stopword removal) on the low-cost implementations of these techniques using a large set (n = 66) of detection models. Our results show the limited impact of preprocessing on model performance, with the best results for less noisy data being achieved by neural network- and machine-learning-based models, in contrast to the more robust performance of dictionary-based models on noisy data.

arXiv:2207.00436

Shai-am: A Machine Learning Platform for Investment Strategies

The finance industry has adopted machine learning (ML) as a form of quantitative research to support better investment decisions, yet there are several challenges often overlooked in practice. (1) ML code tends to be unstructured and ad hoc, which hinders cooperation with others. (2) Resource requirements and dependencies vary depending on which algorithm is used, so a flexible and scalable system is needed. (3) It is difficult for domain experts in traditional finance to apply their experience and knowledge in ML-based strategies unless they acquire expertise in recent technologies. This paper presents Shai-am, an ML platform integrated with our own Python framework. The platform leverages existing modern open-source technologies, managing containerized pipelines for ML-based strategies with unified interfaces to solve the aforementioned issues. Each strategy implements the interface defined in the core framework. The framework is designed to enhance reusability and readability, facilitating collaborative work in quantitative research. Shai-am aims to be a pure AI asset manager for solving various tasks in financial markets.

arXiv:2207.00421

Generative Adversarial Networks and Image-Based Malware Classification

For efficient malware removal, determination of malware threat levels, and damage estimation, malware family classification plays a critical role. In this paper, we extract features from malware executable files and represent them as images using various approaches. We then focus on Generative Adversarial Networks (GAN) for multiclass classification and compare our GAN results to other popular machine learning techniques, including Support Vector Machine (SVM), XGBoost, and Restricted Boltzmann Machines (RBM). We find that the AC-GAN discriminator is generally competitive with other machine learning techniques. We also evaluate the utility of the GAN generative model for adversarial attacks on image-based malware detection. While AC-GAN generated images are visually impressive, we find that they are easily distinguished from real malware images using any of several learning techniques. This result indicates that our GAN generated images would be of little value in adversarial attacks.

arXiv:2207.00415

AI in 6G: Energy-Efficient Distributed Machine Learning for Multilayer Heterogeneous Networks

Adept network management is key for supporting extremely heterogeneous applications with stringent quality of service (QoS) requirements; this is more so when envisioning the complex and ultra-dense 6G mobile heterogeneous network (HetNet). From both the environmental and economical perspectives, non-homogeneous QoS demands obstruct the minimization of the energy footprints and operational costs of the envisioned robust networks. As such, network intelligentization is expected to play an essential role in the realization of such sophisticated aims. The fusion of artificial intelligence (AI) and mobile networks will allow for the dynamic and automatic configuration of network functionalities. Machine learning (ML), one of the backbones of AI, will be instrumental in forecasting changes in network loads and resource utilization, estimating channel conditions, optimizing network slicing, and enhancing security and encryption. However, it is well known that ML tasks themselves incur massive computational burdens and energy costs. To overcome such obstacles, we propose a novel layer-based HetNet architecture which optimally distributes tasks associated with different ML approaches across network layers and entities; such a HetNet boasts multiple access schemes as well as device-to-device (D2D) communications to enhance energy efficiency via collaborative learning and communications.

arXiv:2207.00392

Better Methods and Theory for Federated Learning: Compression, Client Selection and Heterogeneity

Federated learning (FL) is an emerging machine learning paradigm involving multiple clients, e.g., mobile phone devices, with an incentive to collaborate in solving a machine learning problem coordinated by a central server. FL was proposed in 2016 by Kone\v{c}n\'{y} et al. and McMahan et al. as a viable privacy-preserving alternative to traditional centralized machine learning since, by construction, the training data points are decentralized and never transferred by the clients to a central server. Therefore, to a certain degree, FL mitigates the privacy risks associated with centralized data collection. Unfortunately, optimization for FL faces several specific issues that centralized optimization usually does not need to handle. In this thesis, we identify several of these challenges and propose new methods and algorithms to address them, with the ultimate goal of enabling practical FL solutions supported with mathematically rigorous guarantees.

arXiv:2207.00283공식 코드 · Papers with Code

Learning Lattice Quantum Field Theories with Equivariant Continuous Flows

We propose a novel machine learning method for sampling from the high-dimensional probability distributions of Lattice Field Theories, which is based on a single neural ODE layer and incorporates the full symmetries of the problem. We test our model on the $\phi^4$ theory, showing that it systematically outperforms previously proposed flow-based methods in sampling efficiency, and the improvement is especially pronounced for larger lattices. Furthermore, we demonstrate that our model can learn a continuous family of theories at once, and the results of learning can be transferred to larger lattices. Such generalizations further accentuate the advantages of machine learning methods.

arXiv:2207.00108

Discrimination in machine learning algorithms

Machine learning algorithms are routinely used for business decisions that may directly affect individuals, for example, because a credit scoring algorithm refuses them a loan. It is then relevant from an ethical (and legal) point of view to ensure that these algorithms do not discriminate based on sensitive attributes (like sex or race), which may occur unwittingly and unknowingly by the operator and the management. Statistical tools and methods are then required to detect and eliminate such potential biases.

arXiv:2207.00300

Robust Bayesian Learning for Reliable Wireless AI: Framework and Applications

This work takes a critical look at the application of conventional machine learning methods to wireless communication problems through the lens of reliability and robustness. Deep learning techniques adopt a frequentist framework, and are known to provide poorly calibrated decisions that do not reproduce the true uncertainty caused by limitations in the size of the training data. Bayesian learning, while in principle capable of addressing this shortcoming, is in practice impaired by model misspecification and by the presence of outliers. Both problems are pervasive in wireless communication settings, in which the capacity of machine learning models is subject to resource constraints and training data is affected by noise and interference. In this context, we explore the application of the framework of robust Bayesian learning. After a tutorial-style introduction to robust Bayesian learning, we showcase the merits of robust Bayesian learning on several important wireless communication problems in terms of accuracy, calibration, and robustness to outliers and misspecification.

arXiv:2207.00091

Threat Assessment in Machine Learning based Systems

Machine learning is a field of artificial intelligence (AI) that is becoming essential for several critical systems, making it a good target for threat actors. Threat actors exploit different Tactics, Techniques, and Procedures (TTPs) against the confidentiality, integrity, and availability of Machine Learning (ML) systems. During the ML cycle, they exploit adversarial TTPs to poison data and fool ML-based systems. In recent years, multiple security practices have been proposed for traditional systems but they are not enough to cope with the nature of ML-based systems. In this paper, we conduct an empirical study of threats reported against ML-based systems with the aim to understand and characterize the nature of ML threats and identify common mitigation strategies. The study is based on 89 real-world ML attack scenarios from the MITRE's ATLAS database, the AI Incident Database, and the literature; 854 ML repositories from the GitHub search and the Python Packaging Advisory database, selected based on their reputation. Attacks from the AI Incident Database and the literature are used to identify vulnerabilities and new types of threats that were not documented in ATLAS. Results show that convolutional neural networks were one of the most targeted models among the attack scenarios. ML repositories with the largest vulnerability prominence include TensorFlow, OpenCV, and Notebook. In this paper, we also report the most frequent vulnerabilities in the studied ML repositories, the most targeted ML phases and models, the most used TTPs in ML phases and attack scenarios. This information is particularly important for red/blue teams to better conduct attacks/defenses, for practitioners to prevent threats during ML development, and for researchers to develop efficient defense mechanisms.

arXiv:2206.14876

AI in Asset Management and Rebellion Research

On October 30th, 2021, Rebellion Research's CEO announced in a Q3 2021 Letter to Investors that Rebellion's AI Global Equity strategy returned +6.8% gross for the first three quarters of 2021. "It's no surprise", Alex told us, "Our Machine Learning global strategy has a history of outperforming the S&P 500 for 14 years". In 2021, Rebellion's brokerage accounts can be opened in over 70 countries, and Rebellion's research covers over 50 countries. Besides being an AI asset management company, Rebellion also defines itself as a top-tier, global machine learning think tank. Alex planned to build a Rebellion ML & AI ecosystem. Should Rebellion stay in the asset management area or jump into other areas? How could the Rebellion strategically move towards a more broad area? What were Rebellion's new or alternative business models?

arXiv:2206.15475

Causal Machine Learning: A Survey and Open Problems

Causal Machine Learning (CausalML) is an umbrella term for machine learning methods that formalize the data-generation process as a structural causal model (SCM). This perspective enables us to reason about the effects of changes to this process (interventions) and what would have happened in hindsight (counterfactuals). We categorize work in CausalML into five groups according to the problems they address: (1) causal supervised learning, (2) causal generative modeling, (3) causal explanations, (4) causal fairness, and (5) causal reinforcement learning. We systematically compare the methods in each category and point out open problems. Further, we review data-modality-specific applications in computer vision, natural language processing, and graph representation learning. Finally, we provide an overview of causal benchmarks and a critical discussion of the state of this nascent field, including recommendations for future work.

arXiv:2206.15303

Physics-informed machine learning for Structural Health Monitoring

The use of machine learning in Structural Health Monitoring is becoming more common, as many of the inherent tasks (such as regression and classification) in developing condition-based assessment fall naturally into its remit. This chapter introduces the concept of physics-informed machine learning, where one adapts ML algorithms to account for the physical insight an engineer will often have of the structure they are attempting to model or assess. The chapter will demonstrate how grey-box models, that combine simple physics-based models with data-driven ones, can improve predictive capability in an SHM setting. A particular strength of the approach demonstrated here is the capacity of the models to generalise, with enhanced predictive capability in different regimes. This is a key issue when life-time assessment is a requirement, or when monitoring data do not span the operational conditions a structure will undergo. The chapter will provide an overview of physics-informed ML, introducing a number of new approaches for grey-box modelling in a Bayesian setting. The main ML tool discussed will be Gaussian process regression, we will demonstrate how physical assumptions/models can be incorporated through constraints, through the mean function and kernel design, and finally in a state-space setting. A range of SHM applications will be demonstrated, from loads monitoring tasks for off-shore and aerospace structures, through to performance monitoring for long-span bridges.

arXiv:2206.15285

Machine learning for automated quality control in injection moulding manufacturing

Machine learning (ML) may improve and automate quality control (QC) in injection moulding manufacturing. As the labelling of extensive, real-world process data is costly, however, the use of simulated process data may offer a first step towards a successful implementation. In this study, simulated data was used to develop a predictive model for the product quality of an injection moulded sorting container. The achieved accuracy, specificity and sensitivity on the test set was $99.4\%$, $99.7\%$ and $94.7\%$, respectively. This study thus shows the potential of ML towards automated QC in injection moulding and encourages the extension to ML models trained on real-world data.

arXiv:2206.15190

Recovering network topology and dynamics via sequence characterization

Sequences arise in many real-world scenarios; thus, identifying the mechanisms behind symbol generation is essential to understanding many complex systems. This paper analyzes sequences generated by agents walking on a networked topology. Given that in many real scenarios, the underlying processes generating the sequence is hidden, we investigate whether the reconstruction of the network via the co-occurrence method is useful to recover both the network topology and agent dynamics generating sequences. We found that the characterization of reconstructed networks provides valuable information regarding the process and topology used to create the sequences. In a machine learning approach considering 16 combinations of network topology and agent dynamics as classes, we obtained an accuracy of 87% with sequences generated with less than 40% of nodes visited. Larger sequences turned out to generate improved machine learning models. Our findings suggest that the proposed methodology could be extended to classify sequences and understand the mechanisms behind sequence generation.

arXiv:2206.15162

Using Person Embedding to Enrich Features and Data Augmentation for Classification

Today, machine learning is applied in almost any field. In machine learning, where there are numerous methods, classification is one of the most basic and crucial ones. Various problems can be solved by classification. The feature selection for model setup is extremely important, and producing new features via feature engineering also has a vital place in the success of the model. In our study, fraud detection classification models are built on a labeled and imbalanced dataset as a case-study. Although it is a natural language processing method, a customer space has been created with word embedding, which has been used in different areas, especially for recommender systems. The customer vectors in the created space are fed to the classification model as a feature. Moreover, to increase the number of positive labels, rows with similar characteristics are re-labeled as positive by using customer similarity determined by embedding. The model in which embedding methods are included in the classification, which provides a better representation of customers, has been compared with other models. Considering the results, it is observed that the customer embedding method had a positive effect on the success of the classification models.