09,24,2021

News Blog Paper China
Using Self-Supervised Co-Training to Improve Facial Representation2021-05-13   ${\displaystyle \cong }$
In this paper, at first, the impact of ImageNet pre-training on Facial Expression Recognition (FER) was tested under different augmentation levels. It could be seen from the results that training from scratch could reach better performance compared to ImageNet fine-tuning at stronger augmentation levels. After that, a framework was proposed for standard Supervised Learning (SL), called Hybrid Learning (HL) which used Self-Supervised co-training with SL in Multi-Task Learning (MTL) manner. Leveraging Self-Supervised Learning (SSL) could gain additional information from input data like spatial information from faces which helped the main SL task. It is been investigated how this method could be used for FER problems with self-supervised pre-tasks such as Jigsaw puzzling and in-painting. The supervised head (SH) was helped by these two methods to lower the error rate under different augmentations and low data regime in the same training settings. The state-of-the-art was reached on AffectNet via two completely different HL methods, without utilizing additional datasets. Moreover, HL's effect was shown on two different facial-related problem, head poses estimation and gender recognition, which concluded to reduce in error rate by up to 9% and 1% respectively. Also, we saw that the HL methods prevented the model from reaching overfitting.
 
From Pixels to Legs: Hierarchical Learning of Quadruped Locomotion2020-11-23   ${\displaystyle \cong }$
Legged robots navigating crowded scenes and complex terrains in the real world are required to execute dynamic leg movements while processing visual input for obstacle avoidance and path planning. We show that a quadruped robot can acquire both of these skills by means of hierarchical reinforcement learning (HRL). By virtue of their hierarchical structure, our policies learn to implicitly break down this joint problem by concurrently learning High Level (HL) and Low Level (LL) neural network policies. These two levels are connected by a low dimensional hidden layer, which we call latent command. HL receives a first-person camera view, whereas LL receives the latent command from HL and the robot's on-board sensors to control its actuators. We train policies to walk in two different environments: a curved cliff and a maze. We show that hierarchical policies can concurrently learn to locomote and navigate in these environments, and show they are more efficient than non-hierarchical neural network policies. This architecture also allows for knowledge reuse across tasks. LL networks trained on one task can be transferred to a new task in a new environment. Finally HL, which processes camera images, can be evaluated at much lower and varying frequencies compared to LL, thus reducing computation times and bandwidth requirements.
 
Hinge-Loss Markov Random Fields and Probabilistic Soft Logic2017-11-16   ${\displaystyle \cong }$
A fundamental challenge in developing high-impact machine learning technologies is balancing the need to model rich, structured domains with the ability to scale to big data. Many important problem areas are both richly structured and large scale, from social and biological networks, to knowledge graphs and the Web, to images, video, and natural language. In this paper, we introduce two new formalisms for modeling structured data, and show that they can both capture rich structure and scale to big data. The first, hinge-loss Markov random fields (HL-MRFs), is a new kind of probabilistic graphical model that generalizes different approaches to convex inference. We unite three approaches from the randomized algorithms, probabilistic graphical models, and fuzzy logic communities, showing that all three lead to the same inference objective. We then define HL-MRFs by generalizing this unified objective. The second new formalism, probabilistic soft logic (PSL), is a probabilistic programming language that makes HL-MRFs easy to define using a syntax based on first-order logic. We introduce an algorithm for inferring most-probable variable assignments (MAP inference) that is much more scalable than general-purpose convex optimization methods, because it uses message passing to take advantage of sparse dependency structures. We then show how to learn the parameters of HL-MRFs. The learned HL-MRFs are as accurate as analogous discrete models, but much more scalable. Together, these algorithms enable HL-MRFs and PSL to model rich, structured data at scales not previously possible.
 
Multi-label classification: do Hamming loss and subset accuracy really conflict with each other?2020-11-16   ${\displaystyle \cong }$
Various evaluation measures have been developed for multi-label classification, including Hamming Loss (HL), Subset Accuracy (SA) and Ranking Loss (RL). However, there is a gap between empirical results and the existing theories: 1) an algorithm often empirically performs well on some measure(s) while poorly on others, while a formal theoretical analysis is lacking; and 2) in small label space cases, the algorithms optimizing HL often have comparable or even better performance on the SA measure than those optimizing SA directly, while existing theoretical results show that SA and HL are conflicting measures. This paper provides an attempt to fill up this gap by analyzing the learning guarantees of the corresponding learning algorithms on both SA and HL measures. We show that when a learning algorithm optimizes HL with its surrogate loss, it enjoys an error bound for the HL measure independent of $c$ (the number of labels), while the bound for the SA measure depends on at most $O(c)$. On the other hand, when directly optimizing SA with its surrogate loss, it has learning guarantees that depend on $O(\sqrt{c})$ for both HL and SA measures. This explains the observation that when the label space is not large, optimizing HL with its surrogate loss can have promising performance for SA. We further show that our techniques are applicable to analyze the learning guarantees of algorithms on other measures, such as RL. Finally, the theoretical analyses are supported by experimental results.
 
Deep Multi-Facial Patches Aggregation Network For Facial Expression Recognition2020-02-20   ${\displaystyle \cong }$
In this paper, we propose an approach for Facial Expressions Recognition (FER) based on a deep multi-facial patches aggregation network. Deep features are learned from facial patches using deep sub-networks and aggregated within one deep architecture for expression classification . Several problems may affect the performance of deep-learning based FER approaches, in particular, the small size of existing FER datasets which might not be sufficient to train large deep learning networks. Moreover, it is extremely time-consuming to collect and annotate a large number of facial images. To account for this, we propose two data augmentation techniques for facial expression generation to expand FER labeled training datasets. We evaluate the proposed framework on three FER datasets. Results show that the proposed approach achieves state-of-art FER deep learning approaches performance when the model is trained and tested on images from the same dataset. Moreover, the proposed data augmentation techniques improve the expression recognition rate, and thus can be a solution for training deep learning FER models using small datasets. The accuracy degrades significantly when testing for dataset bias.
 
Generative Models for Fast Calorimeter Simulation.LHCb case2019-04-06   ${\displaystyle \cong }$
Simulation is one of the key components in high energy physics. Historically it relies on the Monte Carlo methods which require a tremendous amount of computation resources. These methods may have difficulties with the expected High Luminosity Large Hadron Collider (HL LHC) need, so the experiment is in urgent need of new fast simulation techniques. We introduce a new Deep Learning framework based on Generative Adversarial Networks which can be faster than traditional simulation methods by 5 order of magnitude with reasonable simulation accuracy. This approach will allow physicists to produce a big enough amount of simulated data needed by the next HL LHC experiments using limited computing resources.
 
The FaceChannel: A Fast & Furious Deep Neural Network for Facial Expression Recognition2020-09-15   ${\displaystyle \cong }$
Current state-of-the-art models for automatic Facial Expression Recognition (FER) are based on very deep neural networks that are effective but rather expensive to train. Given the dynamic conditions of FER, this characteristic hinders such models of been used as a general affect recognition. In this paper, we address this problem by formalizing the FaceChannel, a light-weight neural network that has much fewer parameters than common deep neural networks. We introduce an inhibitory layer that helps to shape the learning of facial features in the last layer of the network and thus improving performance while reducing the number of trainable parameters. To evaluate our model, we perform a series of experiments on different benchmark datasets and demonstrate how the FaceChannel achieves a comparable, if not better, performance to the current state-of-the-art in FER. Our experiments include cross-dataset analysis, to estimate how our model behaves on different affective recognition conditions. We conclude our paper with an analysis of how FaceChannel learns and adapt the learned facial features towards the different datasets.
 
Auxiliary Sequence Labeling Tasks for Disfluency Detection2020-10-23   ${\displaystyle \cong }$
Detecting disfluencies in spontaneous speech is an important preprocessing step in natural language processing and speech recognition applications. In this paper, we propose a method utilizing named entity recognition (NER) and part-of-speech (POS) as auxiliary sequence labeling (SL) tasks for disfluency detection. First, we show that training a disfluency detection model with auxiliary SL tasks can improve its F-score in disfluency detection. Then, we analyze which auxiliary SL tasks are influential depending on baseline models. Experimental results on the widely used English Switchboard dataset show that our method outperforms the previous state-of-the-art in disfluency detection.
 
The FaceChannel: A Light-weight Deep Neural Network for Facial Expression Recognition2020-04-17   ${\displaystyle \cong }$
Current state-of-the-art models for automatic FER are based on very deep neural networks that are difficult to train. This makes it challenging to adapt these models to changing conditions, a requirement from FER models given the subjective nature of affect perception and understanding. In this paper, we address this problem by formalizing the FaceChannel, a light-weight neural network that has much fewer parameters than common deep neural networks. We perform a series of experiments on different benchmark datasets to demonstrate how the FaceChannel achieves a comparable, if not better, performance, as compared to the current state-of-the-art in FER.
 
Facial Emotion Recognition: State of the Art Performance on FER20132021-05-08   ${\displaystyle \cong }$
Facial emotion recognition (FER) is significant for human-computer interaction such as clinical practice and behavioral description. Accurate and robust FER by computer models remains challenging due to the heterogeneity of human faces and variations in images such as different facial pose and lighting. Among all techniques for FER, deep learning models, especially Convolutional Neural Networks (CNNs) have shown great potential due to their powerful automatic feature extraction and computational efficiency. In this work, we achieve the highest single-network classification accuracy on the FER2013 dataset. We adopt the VGGNet architecture, rigorously fine-tune its hyperparameters, and experiment with various optimization methods. To our best knowledge, our model achieves state-of-the-art single-network accuracy of 73.28 % on FER2013 without using extra training data.
 
A Biologically Plausible Supervised Learning Method for Spiking Neural Networks Using the Symmetric STDP Rule2019-10-06   ${\displaystyle \cong }$
Spiking neural networks (SNNs) possess energy-efficient potential due to event-based computation. However, supervised training of SNNs remains a challenge as spike activities are non-differentiable. Previous SNNs training methods can be generally categorized into two basic classes, i.e., backpropagation-like training methods and plasticity-based learning methods. The former methods are dependent on energy-inefficient real-valued computation and non-local transmission, as also required in artificial neural networks (ANNs), whereas the latter are either considered to be biologically implausible or exhibit poor performance. Hence, biologically plausible (bio-plausible) high-performance supervised learning (SL) methods for SNNs remain deficient. In this paper, we proposed a novel bio-plausible SNN model for SL based on the symmetric spike-timing dependent plasticity (sym-STDP) rule found in neuroscience. By combining the sym-STDP rule with bio-plausible synaptic scaling and intrinsic plasticity of the dynamic threshold, our SNN model implemented SL well and achieved good performance in the benchmark recognition task (MNIST dataset). To reveal the underlying mechanism of our SL model, we visualized both layer-based activities and synaptic weights using the t-distributed stochastic neighbor embedding (t-SNE) method after training and found that they were well clustered, thereby demonstrating excellent classification ability. Furthermore, to verify the robustness of our model, we trained it on another more realistic dataset (Fashion-MNIST), which also showed good performance. As the learning rules were bio-plausible and based purely on local spike events, our model could be easily applied to neuromorphic hardware for online training and may be helpful for understanding SL information processing at the synaptic level in biological neural systems.
 
Similarities between policy gradient methods (PGM) in Reinforcement learning (RL) and supervised learning (SL)2019-05-02   ${\displaystyle \cong }$
Reinforcement learning (RL) is about sequential decision making and is traditionally opposed to supervised learning (SL) and unsupervised learning (USL). In RL, given the current state, the agent makes a decision that may influence the next state as opposed to SL (and USL) where, the next state remains the same, regardless of the decisions taken, either in batch or online learning. Although this difference is fundamental between SL and RL, there are connections that have been overlooked. In particular, we prove in this paper that gradient policy method can be cast as a supervised learning problem where true label are replaced with discounted rewards. We provide a new proof of policy gradient methods (PGM) that emphasizes the tight link with the cross entropy and supervised learning. We provide a simple experiment where we interchange label and pseudo rewards. We conclude that other relationships with SL could be made if we modify the reward functions wisely.
 
SL$^2$MF: Predicting Synthetic Lethality in Human Cancers via Logistic Matrix Factorization2018-10-19   ${\displaystyle \cong }$
Synthetic lethality (SL) is a promising concept for novel discovery of anti-cancer drug targets. However, wet-lab experiments for detecting SLs are faced with various challenges, such as high cost, low consistency across platforms or cell lines. Therefore, computational prediction methods are needed to address these issues. This paper proposes a novel SL prediction method, named SL2MF, which employs logistic matrix factorization to learn latent representations of genes from the observed SL data. The probability that two genes are likely to form SL is modeled by the linear combination of gene latent vectors. As known SL pairs are more trustworthy than unknown pairs, we design importance weighting schemes to assign higher importance weights for known SL pairs and lower importance weights for unknown pairs in SL2MF. Moreover, we also incorporate biological knowledge about genes from protein-protein interaction (PPI) data and Gene Ontology (GO). In particular, we calculate the similarity between genes based on their GO annotations and topological properties in the PPI network. Extensive experiments on the SL interaction data from SynLethDB database have been conducted to demonstrate the effectiveness of SL2MF.
 
AI in Pursuit of Happiness, Finding Only Sadness: Multi-Modal Facial Emotion Recognition Challenge2019-10-24   ${\displaystyle \cong }$
The importance of automated Facial Emotion Recognition (FER) grows the more common human-machine interactions become, which will only continue to increase dramatically with time. A common method to describe human sentiment or feeling is the categorical model the `7 basic emotions', consisting of `Angry', `Disgust', `Fear', `Happiness', `Sadness', `Surprise' and `Neutral'. The `Emotion Recognition in the Wild' (EmotiW) competition is now in its 7th year and has become the standard benchmark for measuring FER performance. The focus of this paper is the EmotiW sub-challenge of classifying videos in the `Acted Facial Expression in the Wild' (AFEW) dataset, consisting of both visual and audio modalities, into one of the above classes. Machine learning has exploded as a research topic in recent years, with advancements in `Deep Learning' a key part of this. Although Deep Learning techniques have been widely applied to the FER task by entrants in previous years, this paper has two main contributions: (i) to apply the latest `state-of-the-art' visual and temporal networks and (ii) exploring various methods of fusing features extracted from the visual and audio elements to enrich the information available to the final model making the prediction. There are a number of complex issues that arise when trying to classify emotions for `in-the-wild' video sequences, which the above two approaches attempt to directly address. There are some positive findings when comparing the results of this paper to past submissions, indicating that further research into the proposed methods and fine-tuning of the models deployed, could result in another step forwards in the field of automated FER.
 
Advancements of federated learning towards privacy preservation: from federated learning to split learning2020-11-25   ${\displaystyle \cong }$
In the distributed collaborative machine learning (DCML) paradigm, federated learning (FL) recently attracted much attention due to its applications in health, finance, and the latest innovations such as industry 4.0 and smart vehicles. FL provides privacy-by-design. It trains a machine learning model collaboratively over several distributed clients (ranging from two to millions) such as mobile phones, without sharing their raw data with any other participant. In practical scenarios, all clients do not have sufficient computing resources (e.g., Internet of Things), the machine learning model has millions of parameters, and its privacy between the server and the clients while training/testing is a prime concern (e.g., rival parties). In this regard, FL is not sufficient, so split learning (SL) is introduced. SL is reliable in these scenarios as it splits a model into multiple portions, distributes them among clients and server, and trains/tests their respective model portions to accomplish the full model training/testing. In SL, the participants do not share both data and their model portions to any other parties, and usually, a smaller network portion is assigned to the clients where data resides. Recently, a hybrid of FL and SL, called splitfed learning, is introduced to elevate the benefits of both FL (faster training/testing time) and SL (model split and training). Following the developments from FL to SL, and considering the importance of SL, this chapter is designed to provide extensive coverage in SL and its variants. The coverage includes fundamentals, existing findings, integration with privacy measures such as differential privacy, open problems, and code implementation.
 
Evaluation and Optimization of Distributed Machine Learning Techniques for Internet of Things2021-03-03   ${\displaystyle \cong }$
Federated learning (FL) and split learning (SL) are state-of-the-art distributed machine learning techniques to enable machine learning training without accessing raw data on clients or end devices. However, their \emph{comparative training performance} under real-world resource-restricted Internet of Things (IoT) device settings, e.g., Raspberry Pi, remains barely studied, which, to our knowledge, have not yet been evaluated and compared, rendering inconvenient reference for practitioners. This work firstly provides empirical comparisons of FL and SL in real-world IoT settings regarding (i) learning performance with heterogeneous data distributions and (ii) on-device execution overhead. Our analyses in this work demonstrate that the learning performance of SL is better than FL under an imbalanced data distribution but worse than FL under an extreme non-IID data distribution. Recently, FL and SL are combined to form splitfed learning (SFL) to leverage each of their benefits (e.g., parallel training of FL and lightweight on-device computation requirement of SL). This work then considers FL, SL, and SFL, and mount them on Raspberry Pi devices to evaluate their performance, including training time, communication overhead, power consumption, and memory usage. Besides evaluations, we apply two optimizations. Firstly, we generalize SFL by carefully examining the possibility of a hybrid type of model training at the server-side. The generalized SFL merges sequential (dependent) and parallel (independent) processes of model training and is thus beneficial for a system with large-scaled IoT devices, specifically at the server-side operations. Secondly, we propose pragmatic techniques to substantially reduce the communication overhead by up to four times for the SL and (generalized) SFL.
 
Symmetric Cross Entropy for Robust Learning with Noisy Labels2019-08-16   ${\displaystyle \cong }$
Training accurate deep neural networks (DNNs) in the presence of noisy labels is an important and challenging task. Though a number of approaches have been proposed for learning with noisy labels, many open issues remain. In this paper, we show that DNN learning with Cross Entropy (CE) exhibits overfitting to noisy labels on some classes ("easy" classes), but more surprisingly, it also suffers from significant under learning on some other classes ("hard" classes). Intuitively, CE requires an extra term to facilitate learning of hard classes, and more importantly, this term should be noise tolerant, so as to avoid overfitting to noisy labels. Inspired by the symmetric KL-divergence, we propose the approach of \textbf{Symmetric cross entropy Learning} (SL), boosting CE symmetrically with a noise robust counterpart Reverse Cross Entropy (RCE). Our proposed SL approach simultaneously addresses both the under learning and overfitting problem of CE in the presence of noisy labels. We provide a theoretical analysis of SL and also empirically show, on a range of benchmark and real-world datasets, that SL outperforms state-of-the-art methods. We also show that SL can be easily incorporated into existing methods in order to further enhance their performance.
 
Graph Generative Models for Fast Detector Simulations in High Energy Physics2021-04-04   ${\displaystyle \cong }$
Accurate and fast simulation of particle physics processes is crucial for the high-energy physics community. Simulating particle interactions with detectors is both time consuming and computationally expensive. With the proton-proton collision energy of 13 TeV, the Large Hadron Collider is uniquely positioned to detect and measure the rare phenomena that can shape our knowledge of new interactions. The High-Luminosity Large Hadron Collider (HL-LHC) upgrade will put a significant strain on the computing infrastructure due to increased event rate and levels of pile-up. Simulation of high-energy physics collisions needs to be significantly faster without sacrificing the physics accuracy. Machine learning approaches can offer faster solutions, while maintaining a high level of fidelity. We discuss a graph generative model that provides effective reconstruction of LHC events, paving the way for full detector level fast simulation for HL-LHC.
 
SplitFed: When Federated Learning Meets Split Learning2020-04-25   ${\displaystyle \cong }$
Federated learning (FL) and split learning (SL) are two recent distributed machine learning (ML) approaches that have gained attention due to their inherent privacy-preserving capabilities. Both approaches follow a model-to-data scenario, in that an ML model is sent to clients for network training and testing. However, FL and SL show contrasting strengths and weaknesses. For example, while FL performs faster than SL due to its parallel client-side model generation strategy, SL provides better privacy than FL due to the split of the ML model architecture between clients and the server. In contrast to FL, SL enables ML training with clients having low computing resources as the client trains only the first few layers of the split ML network model. In this paper, we present a novel approach, named splitfed (SFL), that amalgamates the two approaches eliminating their inherent drawbacks. SFL splits the network architecture between the clients and server as in SL to provide a higher level of privacy than FL. Moreover, it offers better efficiency than SL by incorporating the parallel ML model update paradigm of FL. Our empirical results considering uniformly distributed horizontally partitioned datasets and multiple clients show that SFL provides similar communication efficiency and test accuracies as SL, while significantly reducing - around five times - its computation time per global epoch. Furthermore, as in SL, its communication efficiency over FL improves with the increase in the number of clients.
 
Simple, Fast, Accurate Intent Classification and Slot Labeling for Goal-Oriented Dialogue Systems2019-07-17   ${\displaystyle \cong }$
With the advent of conversational assistants, like Amazon Alexa, Google Now, etc., dialogue systems are gaining a lot of traction, especially in industrial setting. These systems typically consist of Spoken Language understanding component which, in turn, consists of two tasks - Intent Classification (IC) and Slot Labeling (SL). Generally, these two tasks are modeled together jointly to achieve best performance. However, this joint modeling adds to model obfuscation. In this work, we first design framework for a modularization of joint IC-SL task to enhance architecture transparency. Then, we explore a number of self-attention, convolutional, and recurrent models, contributing a large-scale analysis of modeling paradigms for IC+SL across two datasets. Finally, using this framework, we propose a class of 'label-recurrent' models that otherwise non-recurrent, with a 10-dimensional representation of the label history, and show that our proposed systems are easy to interpret, highly accurate (achieving over 30% error reduction in SL over the state-of-the-art on the Snips dataset), as well as fast, at 2x the inference and 2/3 to 1/2 the training time of comparable recurrent models, thus giving an edge in critical real-world systems.