Image captioning models generally lack the capability to take into account user interest, and usually default to global descriptions that try to balance readability, informativeness, and information overload. On the other hand, VQA models generally lack the ability to provide long descriptive answers, while expecting the textual question to be quite precise. We present a method to control the concepts that an image caption should focus on, using an additional input called the guiding text that refers to either groundable or ungroundable concepts in the image. Our model consists of a Transformer-based multimodal encoder that uses the guiding text together with global and object-level image features to derive early-fusion representations used to generate the guided caption. While models trained on Visual Genome data have an in-domain advantage of fitting well when guided with automatic object labels, we find that guided captioning models trained on Conceptual Captions generalize better on out-of-domain images and guiding texts. Our human-evaluation results indicate that attempting in-the-wild guided image captioning requires access to large, unrestricted-domain training datasets, and that increased style diversity (even without increasing the number of unique tokens) is a key factor for improved performance.
Studying the micro-trading behaviors before stock price jumps is an important problem for financial regulations and investment decisions. In this study, we provide a new framework to study pre-jump trading behaviors based on multivariate time series analysis. Different from the existing literature, our methodology takes into account the temporal information embedded in the trading-related attributes and can better evaluate and compare the abnormality levels of different attributes. Moreover, it can explore the joint informativeness of the attributes as well as select a subset of highly informative but minimally redundant attributes to analyze the homogeneous and idiosyncratic patterns in the pre-jump trades of individual stocks. In addition, our analysis involves a set of technical indicators to describe micro-trading behaviors. To illustrate the viability of the proposed methodology, an application case is conducted based on the level-2 data of 189 constituent stocks of the China Security Index 300. The individual and joint informativeness levels of the attributes in predicting price jumps are evaluated and compared. To this end, our experiment provides a set of jump indicators that can represent the pre-jump trading behaviors in the Chinese stock market and have detected some stocks with extremely abnormal pre-jump trades.
Many multimedia applications require precise understanding of the rate-distortion characteristics measured by the function relating visual quality to media attributes, for which we term it the generalized rate-distortion (GRD) function. In this study, we explore the GRD behavior of compressed digital videos in a three-dimensional space of bitrate, resolution, and viewing device/condition. Our analysis on a large-scale video dataset reveals that empirical parametric models are systematically biased while exhaustive search methods require excessive computation time to depict the GRD surfaces. By exploiting the properties that all GRD functions share, we develop an Robust Axial-Monotonic Clough-Tocher (RAMCT) interpolation method to model the GRD function. This model allows us to accurately reconstruct the complete GRD function of a source video content from a moderate number of measurements. To further reduce the computational cost, we present a novel sampling scheme based on a probabilistic model and an information measure. The proposed sampling method constructs a sequence of quality queries by minimizing the overall informativeness in the remaining samples. Experimental results show that the proposed algorithm significantly outperforms state-of-the-art approaches in accuracy and efficiency. Finally, we demonstrate the usage of the proposed model in three applications: rate-distortion curve prediction, per-title encoding profile generation, and video encoder comparison.
Recent deep learning methods for recommendation systems are highly sophisticated. For article recommendation task, a neural network encoder which generates a latent representation of the article content would prove useful. However, using raw text with embedding for models could degrade sentence meanings and deteriorate performance. In this paper, we propose PhrecSys (Phrase-based Recommendation System), which injects phrase-level features into content-based recommendation systems to enhance feature informativeness and model interpretability. Experiments conducted on six months of real-world data demonstrate that phrase features boost content-based models in predicting both user click and view behavior. Furthermore, the attention mechanism illustrates that phrase awareness benefits the learning of textual focus by putting the model's attention on meaningful text spans, which leads to interpretable article recommendation.
Filtering relevant documents with respect to entities is an essential task in the context of knowledge base construction and maintenance. It entails processing a time-ordered stream of documents that might be relevant to an entity in order to select only those that contain vital information. State-of-the-art approaches to document filtering for popular entities are entity-dependent: they rely on and are also trained on the specifics of differentiating features for each specific entity. Moreover, these approaches tend to use so-called extrinsic information such as Wikipedia page views and related entities which is typically only available only for popular head entities. Entity-dependent approaches based on such signals are therefore ill-suited as filtering methods for long-tail entities. In this paper we propose a document filtering method for long-tail entities that is entity-independent and thus also generalizes to unseen or rarely seen entities. It is based on intrinsic features, i.e., features that are derived from the documents in which the entities are mentioned. We propose a set of features that capture informativeness, entity-saliency, and timeliness. In particular, we introduce features based on entity aspect similarities, relation patterns, and temporal expressions and combine these with standard features for document filtering. Experiments following the TREC KBA 2014 setup on a publicly available dataset show that our model is able to improve the filtering performance for long-tail entities over several baselines. Results of applying the model to unseen entities are promising, indicating that the model is able to learn the general characteristics of a vital document. The overall performance across all entities---i.e., not just long-tail entities---improves upon the state-of-the-art without depending on any entity-specific training data.
Objective: Automatic text summarization tools can help users in the biomedical domain to access information efficiently from a large volume of scientific literature and other sources of text documents. In this paper, we propose a summarization method that combines itemset mining and domain knowledge to construct a concept-based model and to extract the main subtopics from an input document. Our summarizer quantifies the informativeness of each sentence using the support values of itemsets appearing in the sentence. Methods: To address the concept-level analysis of text, our method initially maps the original document to biomedical concepts using the UMLS. Then, it discovers the essential subtopics of the text using a data mining technique, namely itemset mining, and constructs the summarization model. The employed itemset mining algorithm extracts a set of frequent itemsets containing correlated and recurrent concepts of the input document. The summarizer selects the most related and informative sentences and generates the final summary. Results: We evaluate the performance of our itemset-based summarizer using the Recall-Oriented Understudy for Gisting Evaluation (ROUGE) metrics, performing a set of experiments. The results show that the itemset-based summarizer performs better than the compared methods. The itemset-based summarizer achieves the best scores for all the assessed ROUGE metrics . Conclusion: Compared to the statistical, similarity, and word frequency methods, the proposed method demonstrates that the summarization model obtained from the concept extraction and itemset mining provides the summarizer with an effective metric for measuring the informative content of sentences. This can lead to an improvement in the performance of biomedical literature summarization.
Sensory neurons give highly variable responses to stimulation, which can limit the amount of stimulus information available to downstream circuits. Much work has investigated the factors that affect the amount of information encoded in these population responses, leading to insights about the role of covariability among neurons, tuning curve shape, etc. However, the informativeness of neural responses is not the only relevant feature of population codes; of potentially equal importance is how robustly that information propagates to downstream structures. For instance, to quantify the retina's performance, one must consider not only the informativeness of the optic nerve responses, but also the amount of information that survives the spike-generating nonlinearity and noise corruption in the next stage of processing, the lateral geniculate nucleus. Our study identifies the set of covariance structures for the upstream cells that optimize the ability of information to propagate through noisy, nonlinear circuits. Within this optimal family are covariances with "differential correlations", which are known to reduce the information encoded in neural population activities. Thus, covariance structures that maximize information in neural population codes, and those that maximize the ability of this information to propagate, can be very different.
In this projecting work we propose a mass spectrometric patch-clamp equipment with the capillary performing both a local potential registration at the cell membrane and the analyte suction simultaneously. This paper provides a current literature analysis comparing the possibilities of the novel approach proposed with the known methods, such as scanning patch-clamp, scanning ion conductance microscopy, patch clamp based on scanning probe microscopy technology, quantitative subcellular secondary ion mass spectrometry or "ion microscopy", live single-cell mass spectrometry, in situ cell-by-cell imaging, single-cell video-mass spectrometry, etc. We also consider the ways to improve the informativeness of these methods and particularly emphasize the trend at the increasing of the analysis complexity. We propose here the way to improve the efficiency of the cell trapping to the capillary during MS-path-clamp, as well as to provide laser surface ionization using laser trapping and tweezing of cells with the laser beam transmitted through the capillary as a waveguide. It is also possible to combine the above system with the microcolumn separation system or capillary electrophoresis as an optional direction of further development of the complex of analytical techniques emerging from the MS variation of patch-clamp.
During a disaster event, images shared on social media helps crisis managers gain situational awareness and assess incurred damages, among other response tasks. Recent advances in computer vision and deep neural networks have enabled the development of models for real-time image classification for a number of tasks, including detecting crisis incidents, filtering irrelevant images, classifying images into specific humanitarian categories, and assessing the severity of damage. Despite several efforts, past works mainly suffer from limited resources (i.e., labeled images) available to train more robust deep learning models. In this study, we propose new datasets for disaster type detection, and informativeness classification, and damage severity assessment. Moreover, we relabel existing publicly available datasets for new tasks. We identify exact- and near-duplicates to form non-overlapping data splits, and finally consolidate them to create larger datasets. In our extensive experiments, we benchmark several state-of-the-art deep learning models and achieve promising results. We release our datasets and models publicly, aiming to provide proper baselines as well as to spur further research in the crisis informatics community.
Deep pre-trained Transformer models have achieved state-of-the-art results over a variety of natural language processing (NLP) tasks. By learning rich language knowledge with millions of parameters, these models are usually overparameterized and significantly increase the computational overhead in applications. It is intuitive to address this issue by model compression. In this work, we propose a method, called Single-Shot Meta-Pruning, to compress deep pre-trained Transformers before fine-tuning. Specifically, we focus on pruning unnecessary attention heads adaptively for different downstream tasks. To measure the informativeness of attention heads, we train our Single-Shot Meta-Pruner (SMP) with a meta-learning paradigm aiming to maintain the distribution of text representations after pruning. Compared with existing compression methods for pre-trained models, our method can reduce the overhead of both fine-tuning and inference. Experimental results show that our pruner can selectively prune 50% of attention heads with little impact on the performance on downstream tasks and even provide better text representations. The source code will be released in the future.
Active learning effectively collects data instances for training deep learning models when the labeled dataset is limited and the annotation cost is high. Besides active learning, data augmentation is also an effective technique to enlarge the limited amount of labeled instances. However, the potential gain from virtual instances generated by data augmentation has not been considered in the acquisition process of active learning yet. Looking ahead the effect of data augmentation in the process of acquisition would select and generate the data instances that are informative for training the model. Hence, this paper proposes Look-Ahead Data Acquisition via augmentation, or LADA, to integrate data acquisition and data augmentation. LADA considers both 1) unlabeled data instance to be selected and 2) virtual data instance to be generated by data augmentation, in advance of the acquisition process. Moreover, to enhance the informativeness of the virtual data instances, LADA optimizes the data augmentation policy to maximize the predictive acquisition score, resulting in the proposal of InfoMixup and InfoSTN. As LADA is a generalizable framework, we experiment with the various combinations of acquisition and augmentation methods. The performance of LADA shows a significant improvement over the recent augmentation and acquisition baselines which were independently applied to the benchmark datasets.
This paper studies active learning (AL) on graphs, whose purpose is to discover the most informative nodes to maximize the performance of graph neural networks (GNNs). Previously, most graph AL methods focus on learning node representations from a carefully selected labeled dataset with large amount of unlabeled data neglected. Motivated by the success of contrastive learning (CL), we propose a novel paradigm that seamlessly integrates graph AL with CL. While being able to leverage the power of abundant unlabeled data in a self-supervised manner, nodes selected by AL further provide semantic information that can better guide representation learning. Besides, previous work measures the informativeness of nodes without considering the neighborhood propagation scheme of GNNs, so that noisy nodes may be selected. We argue that due to the smoothing nature of GNNs, the central nodes from homophilous subgraphs should benefit the model training most. To this end, we present a minimax selection scheme that explicitly harnesses neighborhood information and discover homophilous subgraphs to facilitate active selection. Comprehensive, confounding-free experiments on five public datasets demonstrate the superiority of our method over state-of-the-arts.
Labeling data correctly is an expensive and challenging task in machine learning, especially for on-line data streams. Deep learning models especially require a large number of clean labeled data that is very difficult to acquire in real-world problems. Choosing useful data samples to label while minimizing the cost of labeling is crucial to maintain efficiency in the training process. When confronted with multiple labelers with different expertise and respective labeling costs, deciding which labeler to choose is nontrivial. In this paper, we consider a novel weak and strong labeler problem inspired by humans natural ability for labeling, in the presence of data streams with noisy labels and constrained by a limited budget. We propose an on-line active learning algorithm that consists of four steps: filtering, adding diversity, informative sample selection, and labeler selection. We aim to filter out the suspicious noisy samples and spend the budget on the diverse informative data using strong and weak labelers in a cost-effective manner. We derive a decision function that measures the information gain by combining the informativeness of individual samples and model confidence. We evaluate our proposed algorithm on the well-known image classification datasets CIFAR10 and CIFAR100 with up to 60% noise. Experiments show that by intelligently deciding which labeler to query, our algorithm maintains the same accuracy compared to the case of having only one of the labelers available while spending less of the budget.
Feature visualizations such as synthetic maximally activating images are a widely used explanation method to better understand the information processing of convolutional neural networks (CNNs). At the same time, there are concerns that these visualizations might not accurately represent CNNs' inner workings. Here, we measure how much extremely activating images help humans to predict CNN activations. Using a well-controlled psychophysical paradigm, we compare the informativeness of synthetic images by Olah et al. (2017) with a simple baseline visualization, namely exemplary natural images that also strongly activate a specific feature map. Given either synthetic or natural reference images, human participants choose which of two query images leads to strong positive activation. The experiments are designed to maximize participants' performance, and are the first to probe intermediate instead of final layer representations. We find that synthetic images indeed provide helpful information about feature map activations ($82\pm4\%$ accuracy; chance would be $50\%$). However, natural images - originally intended as a baseline - outperform synthetic images by a wide margin ($92\pm2\%$). Additionally, participants are faster and more confident for natural images, whereas subjective impressions about the interpretability of the feature visualizations are mixed. The higher informativeness of natural images holds across most layers, for both expert and lay participants as well as for hand- and randomly-picked feature visualizations. Even if only a single reference image is given, synthetic images provide less information than natural images ($65\pm5\%$ vs. $73\pm4\%$). In summary, synthetic images from a popular feature visualization method are significantly less informative for assessing CNN activations than natural images. We argue that visualization methods should improve over this baseline.
State-of-the-art conversational agents have advanced significantly in conjunction with the use of large transformer-based language models. However, even with these advancements, conversational agents still lack the ability to produce responses that are informative and coherent with the local context. In this work, we propose a dialog framework that incorporates both local knowledge as well as users' past dialogues to generate high quality conversations. We introduce an approach to build a dataset based on Reddit conversations, where outbound URL links are widely available in the conversations and the hyperlinked documents can be naturally included as local external knowledge. Using our framework and dataset, we demonstrate that incorporating local knowledge can largely improve informativeness, coherency and realisticness measures using human evaluations. In particular, our approach consistently outperforms the state-of-the-art conversational model on the Reddit dataset across all three measures. We also find that scaling the size of our models from 117M to 8.3B parameters yields consistent improvement of validation perplexity as well as human evaluated metrics. Our model with 8.3B parameters can generate human-like responses as rated by various human evaluations in a single-turn dialog setting.
This paper looks into the problem of handling imbalanced data in a multi-label classification problem. The problem is solved by proposing two novel methods that primarily exploit the geometric relationship between the feature vectors. The first one is an undersampling algorithm that uses angle between feature vectors to select more informative samples while rejecting the less informative ones. A suitable criterion is proposed to define the informativeness of a given sample. The second one is an oversampling algorithm that uses a generative algorithm to create new synthetic data that respects all class boundaries. This is achieved by finding \emph{no man's land} based on Euclidean distance between the feature vectors. The efficacy of the proposed methods is analyzed by solving a generic multi-class recognition problem based on mixture of Gaussians. The superiority of the proposed algorithms is established through comparison with other state-of-the-art methods, including SMOTE and ADASYN, over ten different publicly available datasets exhibiting high-to-extreme data imbalance. These two methods are combined into a single data processing framework and is labeled as ``GICaPS'' to highlight the role of geometry-based information (GI) sampling and Class-Prioritized Synthesis (CaPS) in dealing with multi-class data imbalance problem, thereby making a novel contribution in this field.
In the design optimization of metal forming, it is increasingly significant to use surrogate models to analyse the finite element analysis (FEA) simulations. However, traditional surrogate models using scalar based machine learning methods (SBMLMs) fall in short of accuracy and generalizability. This is because SBMLMs fail to harness the location information of the simulations. To overcome these shortcomings, image based machine learning methods (IBMLMs) are leveraged in this paper. The underlying theory of location information, which supports the advantages of IBMLM, is qualitatively interpreted. Based on this theory, a Res-SE-U-Net IBMLM surrogate model is developed and compared with a multi-layer perceptron (MLP) as a referencing SBMLM surrogate model. It is demonstrated that the IBMLM model is advantageous over the MLP SBMLM model in accuracy, generalizability, robustness, and informativeness. This paper presents a promising methodology of leveraging IBMLMs in surrogate models to make maximum use of info from FEA results. Future prospective studies that inspired by this paper are also discussed.
Sentence function is an important linguistic feature indicating the communicative purpose in uttering a sentence. Incorporating sentence functions into conversations has shown improvements in the quality of generated responses. However, the number of utterances for different types of fine-grained sentence functions is extremely imbalanced. Besides a small number of high-resource sentence functions, a large portion of sentence functions is infrequent. Consequently, dialogue generation conditioned on these infrequent sentence functions suffers from data deficiency. In this paper, we investigate a structured meta-learning (SML) approach for dialogue generation on infrequent sentence functions. We treat dialogue generation conditioned on different sentence functions as separate tasks, and apply model-agnostic meta-learning to high-resource sentence functions data. Furthermore, SML enhances meta-learning effectiveness by promoting knowledge customization among different sentence functions but simultaneously preserving knowledge generalization for similar sentence functions. Experimental results demonstrate that SML not only improves the informativeness and relevance of generated responses, but also can generate responses consistent with the target sentence functions.
Evaluation of a document summarization system has been a critical factor to impact the success of the summarization task. Previous approaches, such as ROUGE, mainly consider the informativeness of the assessed summary and require human-generated references for each test summary. In this work, we propose to evaluate the summary qualities without reference summaries by unsupervised contrastive learning. Specifically, we design a new metric which covers both linguistic qualities and semantic informativeness based on BERT. To learn the metric, for each summary, we construct different types of negative samples with respect to different aspects of the summary qualities, and train our model with a ranking loss. Experiments on Newsroom and CNN/Daily Mail demonstrate that our new evaluation method outperforms other metrics even without reference summaries. Furthermore, we show that our method is general and transferable across datasets.
To model diverse responses for a given post, one promising way is to introduce a latent variable into Seq2Seq models. The latent variable is supposed to capture the discourse-level information and encourage the informativeness of target responses. However, such discourse-level information is often too coarse for the decoder to be utilized. To tackle it, our idea is to transform the coarse-grained discourse-level information into fine-grained word-level information. Specifically, we firstly measure the semantic concentration of corresponding target response on the post words by introducing a fine-grained focus signal. Then, we propose a focus-constrained attention mechanism to take full advantage of focus in well aligning the input to the target response. The experimental results demonstrate that by exploiting the fine-grained signal, our model can generate more diverse and informative responses compared with several state-of-the-art models.