One-hot Vector Research Articles

Towards Fine-Grained HBOE with Rendered Orientation Set and Laplace Smoothing

Human body orientation estimation (HBOE) aims to estimate the orientation of a human body relative to the camera’s frontal view. Despite recent advancements in this field, there still exist limitations in achieving fine-grained results. We identify certain defects and propose corresponding approaches as follows: 1). Existing datasets suffer from non-uniform angle distributions, resulting in sparse image data for certain angles. To provide comprehensive and high-quality data, we introduce RMOS (Rendered Model Orientation Set), a rendered dataset comprising 150K accurately labeled human instances with a wide range of orientations. 2). Directly using one-hot vector as labels may overlook the similarity between angle labels, leading to poor supervision. And converting the predictions from radians to degrees enlarges the regression error. To enhance supervision, we employ Laplace smoothing to vectorize the label, which contains more information. For fine-grained predictions, we adopt weighted Smooth-L1-loss to align predictions with the smoothed-label, thus providing robust supervision. 3). Previous works ignore body-part-specific information, resulting in coarse predictions. By employing local-window self-attention, our model could utilize different body part information for more precise orientation estimations. We validate the effectiveness of our method in the benchmarks with extensive experiments and show that our method outperforms state-of-the-art. Project is available at: https://github.com/Whalesong-zrs/Towards-Fine-grained-HBOE.

Dynamic Weights and Prior Reward in Policy Fusion for Compound Agent Learning

In Deep Reinforcement Learning (DRL) domain, a compound learning task is often decomposed into several sub-tasks in a divide-and-conquer manner, each trained separately and then fused concurrently to achieve the original task, referred to as policy fusion. However, the state-of-the-art (SOTA) policy fusion methods treat the importance of sub-tasks equally throughout the task process, eliminating the possibility of the agent relying on different sub-tasks at various stages. To address this limitation, we propose a generic policy fusion approach, referred to as Policy Fusion Learning with Dynamic Weights and Prior Reward (PFLDWPR), to automate the time-varying selection of sub-tasks. Specifically, PFLDWPR produces a time-varying one-hot vector for sub-tasks to dynamically select a suitable sub-task and mask the rest throughout the entire task process, enabling the fused strategy to optimally guide the agent in executing the compound task. The sub-tasks with the dynamic one-hot vector are then aggregated to obtain the action policy for the original task. Moreover, we collect sub-tasks’s rewards at the pre-training stage as a prior reward, which, along with the current reward, is used to train the policy fusion network. Thus, this approach reduces fusion bias by leveraging prior experience. Experimental results under three popular learning tasks demonstrate that the proposed method significantly improves three SOTA policy fusion methods in terms of task duration, episode reward, and score difference.

Molecular Descriptors, Structure Generation, and Inverse QSAR/QSPR Based on SELFIES.

For inverse QSAR/QSPR in conventional molecular design, several chemical structures must be generated and their molecular descriptors must be calculated. However, there is no one-to-one correspondence between the generated chemical structures and molecular descriptors. In this paper, molecular descriptors, structure generation, and inverse QSAR/QSPR based on self-referencing embedded strings (SELFIES), a 100% robust molecular string representation, are proposed. A one-hot vector is converted from SELFIES to SELFIES descriptors x, and an inverse analysis of the QSAR/QSPR model y = f(x) with the objective variable y and molecular descriptor x is conducted. Thus, x values that achieve a target y value are obtained. Based on these values, SELFIES strings or molecules are generated, meaning that inverse QSAR/QSPR is performed successfully. The SELFIES descriptors and SELFIES-based structure generation are verified using datasets of actual compounds. The successful construction of SELFIES-descriptor-based QSAR/QSPR models with predictive abilities comparable to those of models based on other fingerprints is confirmed. A large number of molecules with one-to-one relationships with the values of the SELFIES descriptors are generated. Furthermore, as a case study of inverse QSAR/QSPR, molecules with target y values are generated successfully. The Python code for the proposed method is available at https://github.com/hkaneko1985/dcekit.

Learning Category Distribution for Text Classification

Label smoothing has a wide range of applications in the machine learning field. Nonetheless, label smoothing only softens the targets by adding a uniform distribution into a one-hot vector, which cannot truthfully reflect the underlying relations among categories. However, learning category relations is of vital importance in many fields such as emotion taxonomy and open set recognition. In this work, we propose a method to obtain the label distribution for each category (category distribution) to reveal category relations. Furthermore, based on the learned category distribution, we calculate new soft targets to improve the performance of model classification. Compared with existing methods, our algorithm can improve neural network models without any side information or additional neural network module by considering category relations. Extensive experiments have been conducted on four original datasets and 10 constructed noisy datasets with three basic neural network models to validate our algorithm. The results demonstrate the effectiveness of our algorithm on the classification task. In addition, three experiments (arrangement, clustering, and similarity) are also conducted to validate the intrinsic quality of the learned category distribution. The results indicate that the learned category distribution can well express underlying relations among categories.

Sound event triage: detecting sound events considering priority of classes

We propose a new task for sound event detection (SED): sound event triage (SET). The goal of SET is to detect an arbitrary number of high-priority event classes while allowing misdetections of low-priority event classes where the priority is given for each event class. In conventional methods of SED for targeting a specific sound event class, it is only possible to give priority to a single event class. Moreover, the level of priority is not adjustable, i.e, the conventional methods can use only types of target event class such as one-hot vector, as inputs. To flexibly control much information on the target event, the proposed SET exploits not only types of target sound but also the extent to which each target sound is detected with priority. To implement the detection of events with priority, we propose class-weighted training, in which loss functions and the network are stochastically weighted by the priority parameter of each class. As this is the first paper on SET, we particularly introduce an implementation of single target SET, which is a subtask of SET. The results of the experiments using the URBAN–SED dataset show that the proposed method of single target SET outperforms the conventional SED method by 8.70, 6.66, and 6.09 percentage points for “air_conditioner,” “car_horn,” and “street_music,” respectively, in terms of the intersection-based F-score. For the average score of classes, the proposed methods increase the intersection-based F-score by up to 3.37 percentage points compared with the conventional SED and other target-class-conditioned models.

Concept-Based Label Distribution Learning for Text Classification

Text classification is a crucial task in data mining and artificial intelligence. In recent years, deep learning-based text classification methods have made great development. The deep learning methods supervise model training by representing a label as a one-hot vector. However, the one-hot label representation cannot adequately reflect the relation between an instance and the labels, as labels are often not completely independent, and the instance may be associated with multiple labels in practice. Simply representing the labels as one-hot vectors leads to overconfidence in the model, making it difficult to distinguish some label confusions. In this paper, we propose a simulated label distribution method based on concepts (SLDC) to tackle this problem. This method captures the overlap between the labels by computing the similarity between an instance and the labels and generates a new simulated label distribution for assisting model training. In particular, we incorporate conceptual information from the knowledge base into the representation of instances and labels to address the surface mismatching problem when instances and labels are compared for similarity. Moreover, to fully use the simulated label distribution and the original label vector, we set up a multi-loss function to supervise the training process. Expensive experiments demonstrate the effectiveness of SLDC on five complex text classification datasets. Further experiments also verify that SLDC is especially helpful for confused datasets.

Label prediction based constrained non-negative matrix factorization for semi-supervised multi-view classification

Semi-supervised multi-view classification can improve the performance by leveraging the information from both labeled and unlabeled data. But it is often a challenge to capture the information from the unlabeled multi-view data. By analyzing the relation between labeled and unlabeled data under multi-view scenario, we propose a novel model with the ability of leveraging the latent label information from the unlabeled data. In our model, a label prediction (LP) term is proposed to jointly obtain the predicted labels of unlabeled data from multiple views. The LP term is integrated into a constrained non-negative matrix factorization based multi-view framework. In this way, the LP and the multi-view representation learning are integrated into one joint learning problem, where they boost each other. Particularly, the predicted label vector is formulated to be the one-hot vector, such that the labels can be obtained directly. Moreover, we propose a new lemma about the gradient of the ℓ2,1 norm in the case of 3-factor matrix decomposition and its corollary about multi-factor matrix decomposition. Based on which, we develop an efficient algorithm and prove its convergence. Experimental results verify that our method can obtain state-of-the-art performance.

Sequential Properties Representation Scheme for Recurrent Neural Network-Based Prediction of Therapeutic Peptides.

The discovery of therapeutic peptides is often accelerated by means of virtual screening supported by machine learning-based predictive models. The predictive performance of such models is sensitive to the choice of data and its representation scheme. While the peptide physicochemical and compositional representations fail to distinguish sequence permutations, the amino acid arrangement within the sequence lacks the important information contained in physicochemical, conformational, topological, and geometrical properties. In this paper, we propose a solution to the identified information gap by implementing a hybrid scheme that complements the best traits from both approaches with the aim of predicting antimicrobial and antiviral activities based on experimental data from DRAMP 2.0, AVPdb, and Uniprot data repositories. Using the Friedman test of statistical significance, we compared our hybrid, sequential properties approach to peptide properties, one-hot vector encoding, and word embedding schemes in the 10-fold cross-validation setting, with respect to the F1 score, Matthews correlation coefficient, geometric mean, recall, and precision evaluation metrics. Moreover, the sequence modeling neural network was employed to gain insight into the synergic effect of both properties- and amino acid order-based predictions. The results suggest that sequential properties significantly (P < 0.01) surpasses the aforementioned state-of-the-art representation schemes. This makes it a strong candidate for increasing the predictive power of screening methods based on machine learning, applicable to any category of peptides.

Collaborative Graph Learning for Session-based Recommendation

Session-based recommendation (SBR) , which mainly relies on a user’s limited interactions with items to generate recommendations, is a widely investigated task. Existing methods often apply RNNs or GNNs to model user’s sequential behavior or transition relationship between items to capture her current preference. For training such models, the supervision signals are merely generated from the sequential interactions inside a session, neglecting the correlations of different sessions, which we argue can provide additional supervisions for learning the item representations. Moreover, previous methods mainly adopt the cross-entropy loss for training, where the user’s ground truth preference distribution towards items is regarded as a one-hot vector of the target item, easily making the network over-confident and leading to a serious overfitting problem. Thus, in this article, we propose a Collaborative Graph Learning (CGL) approach for session-based recommendation. CGL first applies the Gated Graph Neural Networks (GGNNs) to learn item embeddings and then is trained by considering both the main supervision as well as the self-supervision signals simultaneously. The main supervisions are produced by the sequential order while the self-supervisions are derived from the global graph constructed by all sessions. In addition, to prevent overfitting, we propose a Target-aware Label Confusion (TLC) learning method in the main supervised component. Extensive experiments are conducted on three publicly available datasets, i.e., Retailrocket, Diginetica, and Gowalla. The experimental results show that CGL can outperform the state-of-the-art baselines in terms of Recall and MRR.

Label Confusion Learning to Enhance Text Classification Models

Representing the true label as one-hot vector is the common practice in training text classification models. However, the one-hot representation may not adequately reflect the relation between the instance and labels, as labels are often not completely independent and instances may relate to multiple labels in practice. The inadequate one-hot representations tend to train the model to be over-confident, which may result in arbitrary prediction and model overfitting, especially for confused datasets (datasets with very similar labels) or noisy datasets (datasets with labeling errors). While training models with label smoothing can ease this problem in some degree, it still fails to capture the realistic relation among labels. In this paper, we propose a novel Label Confusion Model (LCM) as an enhancement component to current popular text classification models. LCM can learn label confusion to capture semantic overlap among labels by calculating the similarity between instance and labels during training and generate a better label distribution to replace the original one-hot label vector, thus improving the final classification performance. Extensive experiments on five text classification benchmark datasets reveal the effectiveness of LCM for several widely used deep learning classification models. Further experiments also verify that LCM is especially helpful for confused or noisy datasets and superior to the label smoothing method.

Zero-Shot Single-Microphone Sound Classification and Localization in a Building Via the Synthesis of Unseen Features

In this paper, we propose a learning-based approach to identify the type and position of sounds using a single microphone in a real-world building. We attempt to treat this problem as a joint classification problem in which we predict the exact positions of sounds while classifying the types that are assumed to be from pre-defined types of sounds. The most problematic issue is that while the types are readily classified under supervised learning frameworks with one-hot encoded labels, it is difficult to predict the exact positions of the sound from unseen positions during training. To address this potential discrepancy, we formulate the position identification problem as a zero-shot learning problem inspired by the human ability to perceive new concepts from previously learned concepts. We extract feature representations from audio data and vectorize the type and position of the sound source as ‘type/position-aware attributes’, instead of labeling each class with a simple one-hot vector. We then train a promising generative model to bridge the extracted features and the attributes by learning the class-invariant structure to transfer the knowledge from seen to unseen classes through their attributes; generative adversarial networks are conditioned on the class-embeddings. Our proposed methods are evaluated on an indoor noise dataset, SNU-B36-EX, a real-world dataset collected inside a building.

Comparison of machine learning and deep learning techniques in promoter prediction across diverse species.

Gene promoters are the key DNA regulatory elements positioned around the transcription start sites and are responsible for regulating gene transcription process. Various alignment-based, signal-based and content-based approaches are reported for the prediction of promoters. However, since all promoter sequences do not show explicit features, the prediction performance of these techniques is poor. Therefore, many machine learning and deep learning models have been proposed for promoter prediction. In this work, we studied methods for vector encoding and promoter classification using genome sequences of three distinct higher eukaryotes viz. yeast (Saccharomyces cerevisiae), A. thaliana (plant) and human (Homo sapiens). We compared one-hot vector encoding method with frequency-based tokenization (FBT) for data pre-processing on 1-D Convolutional Neural Network (CNN) model. We found that FBT gives a shorter input dimension reducing the training time without affecting the sensitivity and specificity of classification. We employed the deep learning techniques, mainly CNN and recurrent neural network with Long Short Term Memory (LSTM) and random forest (RF) classifier for promoter classification at k-mer sizes of 2, 4 and 8. We found CNN to be superior in classification of promoters from non-promoter sequences (binary classification) as well as species-specific classification of promoter sequences (multiclass classification). In summary, the contribution of this work lies in the use of synthetic shuffled negative dataset and frequency-based tokenization for pre-processing. This study provides a comprehensive and generic framework for classification tasks in genomic applications and can be extended to various classification problems.

Multi-style image transfer system using conditional cycleGAN

ABSTRACT This paper aims to extend the capability of Cycle-Consistent Adversarial Network (CycleGAN) by equipping it with a conditional constraint and extend it into a multi-style image transfer system that can transfer images among more than two image domains. The conditional constraint is given in the form of the target style feature map instead of a one-hot vector, and has shown to provide better transfer results. The proposed system offers greater flexibility for users to choose the style for image transfer. Experimental results show that such an architecture is not only feasible but also yields good results. The proposed architecture can be extended to other transformation applications, such as facial expressions transfer, face aging, and synthesis of various features.

Categorizing Malware via A Word2Vec-based Temporal Convolutional Network Scheme

As edge computing paradigm achieves great popularity in recent years, there remain some technical challenges that must be addressed to guarantee smart device security in Internet of Things (IoT) environment. Generally, smart devices transmit individual data across the IoT for various purposes nowadays, and it will cause losses and impose a huge threat to users since malware may steal and damage these data. To improve malware detection performance on IoT smart devices, we conduct a malware categorization analysis based on the Kaggle competition of Microsoft Malware Classification Challenge (BIG 2015) dataset in this article. Practically speaking, motivated by temporal convolutional network (TCN) structure, we propose a malware categorization scheme mainly using Word2Vec pre-trained model. Considering that the popular one-hot encoding converts input names from malicious files to high-dimensional vectors since each name is represented as one dimension in one-hot vector space, more compact vectors with fewer dimensions are obtained through the use of Word2Vec pre-training strategy, and then it can lead to fewer parameters and stronger malware feature representation. Moreover, compared with long short-term memory (LSTM), TCN demonstrates better performance with longer effective memory and faster training speed in sequence modeling tasks. The experimental comparisons on this malware dataset reveal better categorization performance with less memory usage and training time. Especially, through the performance comparison between our scheme and the state-of-the-art Word2Vec-based LSTM approach, our scheme shows approximately 1.3% higher predicted accuracy than the latter on this malware categorization task. Additionally, it also demonstrates that our scheme reduces about 90 thousand parameters and more than 1 hour on the model training time in this comparison.

Modeling Feature Representations for Affective Speech Using Generative Adversarial Networks

Emotion recognition is a classic field of research with a typical setup extracting features and feeding them through a classifier for prediction. On the other hand, generative models jointly capture the distributional relationship between emotions and the feature profiles. Recently, Generative Adversarial Networks (GANs) have surfaced as a new class of generative models and have shown considerable success in modeling distributions in the fields of computer vision and natural language understanding. In this article, we experiment with variants of GAN architectures to generate feature vectors corresponding to an emotion in two ways: (i) A generator is trained with samples from a mixture prior. Each mixture component corresponds to an emotional class and can be sampled to generate features from the corresponding emotion. (ii) A one-hot vector corresponding to an emotion can be explicitly used to generate the features. We perform analysis on such models and also propose different metrics used to measure the performance of the GAN models in their ability to generate realistic synthetic samples. Apart from evaluation on a given dataset of interest, we perform a cross-corpus study where we study the utility of the synthetic samples as additional training data in low resource conditions.

Attribute-based structural damage identification by few-shot meta learning with inter-class knowledge transfer

Image archives of multi-class structural damages can be collected by manual inspection and then used for structural damage identification. On one hand, conventional image-processing-based approaches rely on optimal designs of hand-crafted feature detectors and lack universal adaptability for various application cases; on the other hand, regular supervised learning techniques require complete damage types and sufficient training examples to establish a robust damage recognition model, which brings up a time-labor-consuming image collection process. To solve these problems, this study proposes a nested attribute-based few-shot meta learning paradigm for structural damage identification. First, an external few-shot meta learning module is established based on different classification tasks named as meta-batches to produce robust classifiers for new damage types, in which support and query subsets including partial damage types and a few examples are randomly sampled from the original image dataset. Second, an embedded internal attribute-based transfer learning model is trained by minimizing the l2-norm and angular losses of attribute representation vectors in an end-to-end manner, where damage attributes act as the common inter-class knowledge and are transferred from the source damage space of support set to the target damage space of query set. Finally, the proposed approach is validated on a real-world structural damage image dataset, which contains 1000 examples of 10 representative damage types in total. Results show the proposed approach produces an overall accuracy of 93.5% and an average area under the ROC curve of 0.96 for 10 damage types. The general equilibrium of average precision and recall indicates that the trained model is balanced to both positive and negative examples for each damage type. Compared with a regular supervised learning model by directly classifying input images with one-hot vector labels, the proposed approach generates higher accuracy and better robustness. Parameter study suggests the proposed paradigm enables to train a stable and reliable meta learning classification model that can perform well across a series of settings about the ratio between support and query subsets. Theoretical analysis is also performed to explain why meta learning surpasses regular supervised learning.

Pseudo distribution on unseen classes for generalized zero shot learning

Although Zero Shot Learning (ZSL) has attracted more and more attention due to its powerful ability of recognizing new objects without retraining, it has a serious drawback that it only focuses on unseen classes during prediction. To solve this issue, Generalized ZSL (GZSL) extends the search range to both seen and unseen classes, which makes it a more realistic and challenging task. Conventional methods on GZSL often suffer from the domain shift problem on seen classes because they have only seen data for training. Deep Calibration Network (DCN) tries to minimize the entropy of assigning seen data to unseen classes to balance the training on both seen and unseen classes. However, there are still two problems for DCN, one is the hubness problem and another is the lack of training guidance. In this paper, to solve the two problems, we propose a novel method called PSeudo Distribution (PSD), which exploits the attribute similarity between seen classes and unseen classes as the training guidance to assign the seen data to unseen classes. In addition, the attribute similarity is also compressed to one-hot vector to further encourage the certainty of the model. Besides, the visual space is utilized as the embedding space, which can well settle the hubness problem. Extensive experiments are conducted on four popular datasets, and the results show the superiority of the proposed method.

Data-Rate Driven Transmission Strategies for Deep Learning-Based Communication Systems

Deep learning (DL) based autoencoder is a promising architecture to implement end-to-end communication systems. One fundamental problem of such systems is how to increase the transmission rate. Two new schemes are proposed to address the limited data rate issue: adaptive transmission scheme and generalized data representation (GDR) scheme. In the first scheme, an adaptive transmission is designed to select the transmission vectors for maximizing the data rate under different channel conditions. The block error rate (BLER) of the first scheme is 80% lower than that of the conventional one-hot vector scheme. This implies that higher data rate can be achieved by the adaptive transmission scheme. In the second scheme, the GDR replaces the conventional one-hot representation. The GDR scheme can achieve higher data rate than the conventional one-hot vector scheme with comparable BLER performance. For example, when the vector size is eight, the proposed GDR scheme can double the date rate of the one-hot vector scheme. Besides, the joint scheme of the two proposed schemes can create further benefits. The effect of signal-to-noise ratio (SNR) is analyzed for these DL-based communication systems. Numerical results show that training the autoencoder using data set with various SNR values can attain robust BLER performance under different channel conditions.

Prior Knowledge Driven Label Embedding for Slot Filling in Natural Language Understanding

Traditional slot filling in natural language understanding (NLU) predicts a one-hot vector for each word. This form of label representation lacks semantic correlation modelling, which leads to severe data sparsity problem, especially when adapting an NLU model to a new domain. To address this issue, a novel label embedding based slot filling framework is proposed in this paper. Here, distributed label embedding is constructed for each slot using prior knowledge. Three encoding methods are investigated to incorporate different kinds of prior knowledge about slots: atomic concepts, slot descriptions, and slot exemplars. The proposed label embeddings tend to share text patterns and reuses data with different slot labels. This makes it useful for adaptive NLU with limited data. Also, since label embedding is independent of NLU model, it is compatible with almost all deep learning based slot filling models. The proposed approaches are evaluated on three datasets. Experiments on single domain and domain adaptation tasks show that label embedding achieves significant performance improvement over traditional one-hot label representation as well as advanced zero-shot approaches.

End-to-End Autonomous Driving: An Angle Branched Network Approach

Imitation learning for the end-to-end autonomous driving has drawn renewed attention from academic communities. Current methods either only use images as the input, which will yield ambiguities when a vehicle approaches an intersection, or use additional command information to navigate the vehicle but inefficiently. Focusing on making the vehicle automatically drive along the given path, we propose a new and effective navigation command called as subgoal angle which does not require human participation and is calculated by the current position and subgoal of the ego-vehicle. Thus, the subgoal angle contains more information than the navigation command represented as a one-hot vector. Additionally, we propose a model architecture called as angle branched network that makes predictions based on the subgoal angle. In this network, the subgoal angle is not only used for extracting useful features but also for guiding the appropriate prediction layer to make predictions for both the steer angle and the throttle status (which controls the acceleration). Experiments are conducted in a three-dimensional urban simulator. Both quantitive and qualitive results show the effectiveness of the navigation command and the angle branched network. Moreover, the performance can be further boosted by integrating both semantic and depth information into the driving model. Especially by using the depth information, collisions with vehicles and pedestrians can be reduced.