PyTorchKR Papers

최신 논문

총 1,314편
태그·기간 등 상세 필터

카테고리 탐색

로그인 후 이 태그 구독

arXiv:1911.03219

Language Grounding through Social Interactions and Curiosity-Driven Multi-Goal Learning

Autonomous reinforcement learning agents, like children, do not have access to predefined goals and reward functions. They must discover potential goals, learn their own reward functions and engage in their own learning trajectory. Children, however, benefit from exposure to language, helping to organize and mediate their thought. We propose LE2 (Language Enhanced Exploration), a learning algorithm leveraging intrinsic motivations and natural language (NL) interactions with a descriptive social partner (SP). Using NL descriptions from the SP, it can learn an NL-conditioned reward function to formulate goals for intrinsically motivated goal exploration and learn a goal-conditioned policy. By exploring, collecting descriptions from the SP and jointly learning the reward function and the policy, the agent grounds NL descriptions into real behavioral goals. From simple goals discovered early to more complex goals discovered by experimenting on simpler ones, our agent autonomously builds its own behavioral repertoire. This naturally occurring curriculum is supplemented by an active learning curriculum resulting from the agent's intrinsic motivations. Experiments are presented with a simulated robotic arm that interacts with several objects including tools.

arXiv:1911.02357

Uninformed Students: Student-Teacher Anomaly Detection with Discriminative Latent Embeddings

We introduce a powerful student-teacher framework for the challenging problem of unsupervised anomaly detection and pixel-precise anomaly segmentation in high-resolution images. Student networks are trained to regress the output of a descriptive teacher network that was pretrained on a large dataset of patches from natural images. This circumvents the need for prior data annotation. Anomalies are detected when the outputs of the student networks differ from that of the teacher network. This happens when they fail to generalize outside the manifold of anomaly-free training data. The intrinsic uncertainty in the student networks is used as an additional scoring function that indicates anomalies. We compare our method to a large number of existing deep learning based methods for unsupervised anomaly detection. Our experiments demonstrate improvements over state-of-the-art methods on a number of real-world datasets, including the recently introduced MVTec Anomaly Detection dataset that was specifically designed to benchmark anomaly segmentation algorithms.

arXiv:1910.14539

Naver Labs Europe's Systems for the Document-Level Generation and Translation Task at WNGT 2019

Recently, neural models led to significant improvements in both machine translation (MT) and natural language generation tasks (NLG). However, generation of long descriptive summaries conditioned on structured data remains an open challenge. Likewise, MT that goes beyond sentence-level context is still an open issue (e.g., document-level MT or MT with metadata). To address these challenges, we propose to leverage data from both tasks and do transfer learning between MT, NLG, and MT with source-side metadata (MT+NLG). First, we train document-based MT systems with large amounts of parallel data. Then, we adapt these models to pure NLG and MT+NLG tasks by fine-tuning with smaller amounts of domain-specific data. This end-to-end NLG approach, without data selection and planning, outperforms the previous state of the art on the Rotowire NLG task. We participated to the "Document Generation and Translation" task at WNGT 2019, and ranked first in all tracks.

arXiv:1910.12589

Forecasting the Success of Television Series using Machine Learning

Television is an ever-evolving multi billion dollar industry. The success of a television show in an increasingly technological society is a vast multi-variable formula. The art of success is not just something that happens, but is studied, replicated, and applied. Hollywood can be unpredictable regarding success, as many movies and sitcoms that are hyped up and promise to be a hit end up being box office failures and complete disappointments. In current studies, linguistic exploration is being performed on the relationship between Television series and target community of viewers. Having a decision support system that can display sound and predictable results would be needed to build confidence in the investment of a new TV series. The models presented in this study use data to study and determine what makes a sitcom successful. In this paper, we use descriptive and predictive modeling techniques to assess the continuing success of television comedies: The Office, Big Bang Theory, Arrested Development, Scrubs, and South Park. The factors that are tested for statistical significance on episode ratings are character presence, director, and writer. These statistics show that while characters are indeed crucial to the shows themselves, the creation and direction of the shows pose implication upon the ratings and therefore the success of the shows. We use machine learning based forecasting models to accurately predict the success of shows. The models represent a baseline to understanding the success of a television show and how producers can increase the success of current television shows or utilize this data in the creation of future shows. Due to the many factors that go into a series, the empirical analysis in this work shows that there is no one-fits-all model to forecast the rating or success of a television show.

arXiv:1910.11540

Descriptive Dimensionality and Its Characterization of MDL-based Learning and Change Detection

This paper introduces a new notion of dimensionality of probabilistic models from an information-theoretic view point. We call it the "descriptive dimension"(Ddim). We show that Ddim coincides with the number of independent parameters for the parametric class, and can further be extended to real-valued dimensionality when a number of models are mixed. The paper then derives the rate of convergence of the MDL (Minimum Description Length) learning algorithm which outputs a normalized maximum likelihood (NML) distribution with model of the shortest NML codelength. The paper proves that the rate is governed by Ddim. The paper also derives error probabilities of the MDL-based test for multiple model change detection. It proves that they are also governed by Ddim. Through the analysis, we demonstrate that Ddim is an intrinsic quantity which characterizes the performance of the MDL-based learning and change detection.

arXiv:1910.09956

Artificial Intelligence and the Future of Psychiatry: Qualitative Findings from a Global Physician Survey

The potential for machine learning to disrupt the medical profession is the subject of ongoing debate within biomedical informatics. This study aimed to explore psychiatrists' opinions about the potential impact of innovations in artificial intelligence and machine learning on psychiatric practice. In Spring 2019, we conducted a web-based survey of 791 psychiatrists from 22 countries worldwide. The survey measured opinions about the likelihood future technology would fully replace physicians in performing ten key psychiatric tasks. This study involved qualitative descriptive analysis of written response to three open-ended questions in the survey. Comments were classified into four major categories in relation to the impact of future technology on patient-psychiatric interactions, the quality of patient medical care, the profession of psychiatry, and health systems. Overwhelmingly, psychiatrists were skeptical that technology could fully replace human empathy. Many predicted that 'man and machine' would increasingly collaborate in undertaking clinical decisions, with mixed opinions about the benefits and harms of such an arrangement. Participants were optimistic that technology might improve efficiencies and access to care, and reduce costs. Ethical and regulatory considerations received limited attention. This study presents timely information of psychiatrists' view about the scope of artificial intelligence and machine learning on psychiatric practice. Psychiatrists expressed divergent views about the value and impact of future technology with worrying omissions about practice guidelines, and ethical and regulatory issues.

arXiv:1910.08657

Temporal Network Sampling

Temporal networks representing a stream of timestamped edges are seemingly ubiquitous in the real-world. However, the massive size and continuous nature of these networks make them fundamentally challenging to analyze and leverage for descriptive and predictive modeling tasks. In this work, we propose a general framework for temporal network sampling with unbiased estimation. We develop online, single-pass sampling algorithms and unbiased estimators for temporal network sampling. The proposed algorithms enable fast, accurate, and memory-efficient statistical estimation of temporal network patterns and properties. In addition, we propose a temporally decaying sampling algorithm with unbiased estimators for studying networks that evolve in continuous time, where the strength of links is a function of time, and the motif patterns are temporally-weighted. In contrast to the prior notion of a $\bigtriangleup t$-temporal motif, the proposed formulation and algorithms for counting temporally weighted motifs are useful for forecasting tasks in networks such as predicting future links, or a future time-series variable of nodes and links. Finally, extensive experiments on a variety of temporal networks from different domains demonstrate the effectiveness of the proposed algorithms. A detailed ablation study is provided to understand the impact of the various components of the proposed framework.

arXiv:1910.09246

Who wants accurate models? Arguing for a different metrics to take classification models seriously

With the increasing availability of AI-based decision support, there is an increasing need for their certification by both AI manufacturers and notified bodies, as well as the pragmatic (real-world) validation of these systems. Therefore, there is the need for meaningful and informative ways to assess the performance of AI systems in clinical practice. Common metrics (like accuracy scores and areas under the ROC curve) have known problems and they do not take into account important information about the preferences of clinicians and the needs of their specialist practice, like the likelihood and impact of errors and the complexity of cases. In this paper, we present a new accuracy measure, the H-accuracy (Ha), which we claim is more informative in the medical domain (and others of similar needs) for the elements it encompasses. We also provide proof that the H-accuracy is a generalization of the balanced accuracy and establish a relation between the H-accuracy and the Net Benefit. Finally, we illustrate an experimentation in two user studies to show the descriptive power of the Ha score and how complementary and differently informative measures can be derived from its formulation (a Python script to compute Ha is also made available).

arXiv:1910.06624

Probability Logic

This chapter presents probability logic as a rationality framework for human reasoning under uncertainty. Selected formal-normative aspects of probability logic are discussed in the light of experimental evidence. Specifically, probability logic is characterized as a generalization of bivalent truth-functional propositional logic (short "logic"), as being connexive, and as being nonmonotonic. The chapter discusses selected argument forms and associated uncertainty propagation rules. Throughout the chapter, the descriptive validity of probability logic is compared to logic, which was used as the gold standard of reference for assessing the rationality of human reasoning in the 20th century.

arXiv:1910.06136

Component Mismatches Are a Critical Bottleneck to Fielding AI-Enabled Systems in the Public Sector

The use of machine learning or artificial intelligence (ML/AI) holds substantial potential toward improving many functions and needs of the public sector. In practice however, integrating ML/AI components into public sector applications is severely limited not only by the fragility of these components and their algorithms, but also because of mismatches between components of ML-enabled systems. For example, if an ML model is trained on data that is different from data in the operational environment, field performance of the ML component will be dramatically reduced. Separate from software engineering considerations, the expertise needed to field an ML/AI component within a system frequently comes from outside software engineering. As a result, assumptions and even descriptive language used by practitioners from these different disciplines can exacerbate other challenges to integrating ML/AI components into larger systems. We are investigating classes of mismatches in ML/AI systems integration, to identify the implicit assumptions made by practitioners in different fields (data scientists, software engineers, operations staff) and find ways to communicate the appropriate information explicitly. We will discuss a few categories of mismatch, and provide examples from each class. To enable ML/AI components to be fielded in a meaningful way, we will need to understand the mismatches that exist and develop practices to mitigate the impacts of these mismatches.

arXiv:1910.01442

CLEVRER: CoLlision Events for Video REpresentation and Reasoning

The ability to reason about temporal and causal events from videos lies at the core of human intelligence. Most video reasoning benchmarks, however, focus on pattern recognition from complex visual and language input, instead of on causal structure. We study the complementary problem, exploring the temporal and causal structures behind videos of objects with simple visual appearance. To this end, we introduce the CoLlision Events for Video REpresentation and Reasoning (CLEVRER), a diagnostic video dataset for systematic evaluation of computational models on a wide range of reasoning tasks. Motivated by the theory of human casual judgment, CLEVRER includes four types of questions: descriptive (e.g., "what color"), explanatory ("what is responsible for"), predictive ("what will happen next"), and counterfactual ("what if"). We evaluate various state-of-the-art models for visual reasoning on our benchmark. While these models thrive on the perception-based task (descriptive), they perform poorly on the causal tasks (explanatory, predictive and counterfactual), suggesting that a principled approach for causal reasoning should incorporate the capability of both perceiving complex visual and language inputs, and understanding the underlying dynamics and causal relations. We also study an oracle model that explicitly combines these components via symbolic representations.

arXiv:1909.10171공식 코드 · Papers with Code

Syntax-Aware Aspect-Level Sentiment Classification with Proximity-Weighted Convolution Network

It has been widely accepted that Long Short-Term Memory (LSTM) network, coupled with attention mechanism and memory module, is useful for aspect-level sentiment classification. However, existing approaches largely rely on the modelling of semantic relatedness of an aspect with its context words, while to some extent ignore their syntactic dependencies within sentences. Consequently, this may lead to an undesirable result that the aspect attends on contextual words that are descriptive of other aspects. In this paper, we propose a proximity-weighted convolution network to offer an aspect-specific syntax-aware representation of contexts. In particular, two ways of determining proximity weight are explored, namely position proximity and dependency proximity. The representation is primarily abstracted by a bidirectional LSTM architecture and further enhanced by a proximity-weighted convolution. Experiments conducted on the SemEval 2014 benchmark demonstrate the effectiveness of our proposed approach compared with a range of state-of-the-art models.

arXiv:1909.09124

Prediction of overall survival and molecular markers in gliomas via analysis of digital pathology images using deep learning

Cancer histology reveals disease progression and associated molecular processes, and contains rich phenotypic information that is predictive of outcome. In this paper, we developed a computational approach based on deep learning to predict the overall survival and molecular subtypes of glioma patients from microscopic images of tissue biopsies, reflecting measures of microvascular proliferation, mitotic activity, nuclear atypia, and the presence of necrosis. Whole-slide images from 663 unique patients [IDH: 333 IDH-wildtype, 330 IDH-mutants, 1p/19q: 201 1p/19q non-codeleted, 129 1p/19q codeleted] were obtained from TCGA. Sub-images that were free of artifacts and that contained viable tumor with descriptive histologic characteristics were extracted, which were further used for training and testing a deep neural network. The output layer of the network was configured in two different ways: (i) a final Cox model layer to output a prediction of patient risk, and (ii) a final layer with sigmoid activation function, and stochastic gradient decent based optimization with binary cross-entropy loss. Both survival prediction and molecular subtype classification produced promising results using our model. The c-statistic was estimated to be 0.82 (p-value=4.8x10-5) between the risk scores of the proposed deep learning model and overall survival, while accuracies of 88% (area under the curve [AUC]=0.86) were achieved in the detection of IDH mutational status and 1p/19q codeletion. These findings suggest that the deep learning techniques can be applied to microscopic images for objective, accurate, and integrated prediction of outcome for glioma patients. The proposed marker may contribute to (i) stratification of patients into clinical trials, (ii) patient selection for targeted therapy, and (iii) personalized treatment planning.

arXiv:1909.08593공식 코드 · Papers with Code

Fine-Tuning Language Models from Human Preferences

Reward learning enables the application of reinforcement learning (RL) to tasks where reward is defined by human judgment, building a model of reward by asking humans questions. Most work on reward learning has used simulated environments, but complex information about values is often expressed in natural language, and we believe reward learning for language is a key to making RL practical and safe for real-world tasks. In this paper, we build on advances in generative pretraining of language models to apply reward learning to four natural language tasks: continuing text with positive sentiment or physically descriptive language, and summarization tasks on the TL;DR and CNN/Daily Mail datasets. For stylistic continuation we achieve good results with only 5,000 comparisons evaluated by humans. For summarization, models trained with 60,000 comparisons copy whole sentences from the input but skip irrelevant preamble; this leads to reasonable ROUGE scores and very good performance according to our human labelers, but may be exploiting the fact that labelers rely on simple heuristics.

arXiv:1909.08402공식 코드 · Papers with Code

Enriching BERT with Knowledge Graph Embeddings for Document Classification

In this paper, we focus on the classification of books using short descriptive texts (cover blurbs) and additional metadata. Building upon BERT, a deep neural language model, we demonstrate how to combine text representations with metadata and knowledge graph embeddings, which encode author information. Compared to the standard BERT approach we achieve considerably better results for the classification task. For a more coarse-grained classification using eight labels we achieve an F1- score of 87.20, while a detailed classification using 343 labels yields an F1-score of 64.70. We make the source code and trained models of our experiments publicly available

arXiv:1909.07818

Learning Deformable Point Set Registration with Regularized Dynamic Graph CNNs for Large Lung Motion in COPD Patients

Deformable registration continues to be one of the key challenges in medical image analysis. While iconic registration methods have started to benefit from the recent advances in medical deep learning, the same does not yet apply for the registration of point sets, e.g. registration based on surfaces, keypoints or landmarks. This is mainly due to the restriction of the convolution operator in modern CNNs to densely gridded input. However, with the newly developed methods from the field of geometric deep learning suitable tools are now emerging, which enable powerful analysis of medical data on irregular domains. In this work, we present a new method that enables the learning of regularized feature descriptors with dynamic graph CNNs. By incorporating the learned geometric features as prior probabilities into the well-established coherent point drift (CPD) algorithm, formulated as differentiable network layer, we establish an end-to-end framework for robust registration of two point sets. Our approach is evaluated on the challenging task of aligning keypoints extracted from lung CT scans in inhale and exhale states with large deformations and without any additional intensity information. Our results indicate that the inherent geometric structure of the extracted keypoints is sufficient to establish descriptive point features, which yield a significantly improved performance and robustness of our registration framework.

arXiv:1909.07763

A machine vision meta-algorithm for automated recognition of underwater objects using sidescan sonar imagery

This paper details a new method to recognize and detect underwater objects in real-time sidescan sonar data imagery streams, with case-studies of applications for underwater archeology, and ghost fishing gear retrieval. We first synthesize images from sidescan data, apply geometric and radiometric corrections, then use 2D feature detection algorithms to identify point clouds of descriptive visual microfeatures such as corners and edges in the sonar images. We then apply a clustering algorithm on the feature point clouds to group feature sets into regions of interest, reject false positives, yielding a georeferenced inventory of objects.

arXiv:1909.04790

SDM-Net: A Simple and Effective Model for Generalized Zero-Shot Learning

Zero-Shot Learning (ZSL) is a classification task where we do not have even a single training labeled example from a set of unseen classes. Instead, we only have prior information (or description) about seen and unseen classes, often in the form of physically realizable or descriptive attributes. Lack of any single training example from a set of classes prohibits use of standard classification techniques and losses, including the popular crossentropy loss. Currently, state-of-the-art approaches encode the prior class information into dense vectors and optimize some distance between the learned projections of the input vector and the corresponding class vector (collectively known as embedding models). In this paper, we propose a novel architecture of casting zero-shot learning as a standard neural-network with crossentropy loss. During training our approach performs soft-labeling by combining the observed training data for the seen classes with the similarity information from the attributes for which we have no training data or unseen classes. To the best of our knowledge, such similarity based soft-labeling is not explored in the field of deep learning. We evaluate the proposed model on the four benchmark datasets for zero-shot learning, AwA, aPY, SUN and CUB datasets, and show that our model achieves significant improvement over the state-of-the-art methods in Generalized-ZSL and ZSL settings on all of these datasets consistently.

arXiv:1909.02511

CT Data Curation for Liver Patients: Phase Recognition in Dynamic Contrast-Enhanced CT

As the demand for more descriptive machine learning models grows within medical imaging, bottlenecks due to data paucity will exacerbate. Thus, collecting enough large-scale data will require automated tools to harvest data/label pairs from messy and real-world datasets, such as hospital PACS. This is the focus of our work, where we present a principled data curation tool to extract multi-phase CT liver studies and identify each scan's phase from a real-world and heterogenous hospital PACS dataset. Emulating a typical deployment scenario, we first obtain a set of noisy labels from our institutional partners that are text mined using simple rules from DICOM tags. We train a deep learning system, using a customized and streamlined 3D SE architecture, to identify non-contrast, arterial, venous, and delay phase dynamic CT liver scans, filtering out anything else, including other types of liver contrast studies. To exploit as much training data as possible, we also introduce an aggregated cross entropy loss that can learn from scans only identified as "contrast". Extensive experiments on a dataset of 43K scans of 7680 patient imaging studies demonstrate that our 3DSE architecture, armed with our aggregated loss, can achieve a mean F1 of 0.977 and can correctly harvest up to 92.7% of studies, which significantly outperforms the text-mined and standard-loss approach, and also outperforms other, and more complex, model architectures.

arXiv:1909.01497

Iterative Clustering with Game-Theoretic Matching for Robust Multi-consistency Correspondence

Matching corresponding features between two images is a fundamental task to computer vision with numerous applications in object recognition, robotics, and 3D reconstruction. Current state of the art in image feature matching has focused on establishing a single consistency in static scenes; by contrast, finding multiple consistencies in dynamic scenes has been under-researched. In this paper, we present an end-to-end optimization framework named "iterative clustering with Game-Theoretic Matching" (ic-GTM) for robust multi-consistency correspondence. The key idea is to formulate multi-consistency matching as a generalized clustering problem for an image pair. In our formulation, several local matching games are simultaneously carried out in different corresponding block pairs under the guidance of a novel payoff function consisting of both geometric and descriptive compatibility; the global matching results are further iteratively refined by clustering and thresholding with respect to a payoff matrix. We also propose three new metrics for evaluating the performance of multi-consistency image feature matching. Extensive experimental results have shown that the proposed framework significantly outperforms previous state-of-the-art approaches on both singleconsistency and multi-consistency datasets.