Paper Group ANR 809
Accelerating Mini-batch SARAH by Step Size Rules. FairPrep: Promoting Data to a First-Class Citizen in Studies on Fairness-Enhancing Interventions. Real-time texturing for 6D object instance detection from RGB Images. Toward Automated Website Classification by Deep Learning. PT-CoDE: Pre-trained Context-Dependent Encoder for Utterance-level Emotion …
Accelerating Mini-batch SARAH by Step Size Rules
| Title | Accelerating Mini-batch SARAH by Step Size Rules |
| Authors | Zhuang Yang, Zengping Chen, Cheng Wang |
| Abstract | StochAstic Recursive grAdient algoritHm (SARAH), originally proposed for convex optimization and also proven to be effective for general nonconvex optimization, has received great attention due to its simple recursive framework for updating stochastic gradient estimates. The performance of SARAH significantly depends on the choice of step size sequence. However, SARAH and its variants often employ a best-tuned step size by mentor, which is time consuming in practice. Motivated by this gap, we proposed a variant of the Barzilai-Borwein (BB) method, referred to as the Random Barzilai-Borwein (RBB) method, to calculate step size for SARAH in the mini-batch setting, thereby leading to a new SARAH method: MB-SARAH-RBB. We prove that MB-SARAH-RBB converges linearly in expectation for strongly convex objective functions. We analyze the complexity of MB-SARAH-RBB and show that it is better than the original method. Numerical experiments on standard data sets indicate that MB-SARAH-RBB outperforms or matches state-of-the-art algorithms. |
| Tasks | |
| Published | 2019-06-20 |
| URL | https://arxiv.org/abs/1906.08496v1 |
| https://arxiv.org/pdf/1906.08496v1.pdf | |
| PWC | https://paperswithcode.com/paper/accelerating-mini-batch-sarah-by-step-size |
| Repo | |
| Framework | |
FairPrep: Promoting Data to a First-Class Citizen in Studies on Fairness-Enhancing Interventions
| Title | FairPrep: Promoting Data to a First-Class Citizen in Studies on Fairness-Enhancing Interventions |
| Authors | Sebastian Schelter, Yuxuan He, Jatin Khilnani, Julia Stoyanovich |
| Abstract | The importance of incorporating ethics and legal compliance into machine-assisted decision-making is broadly recognized. Further, several lines of recent work have argued that critical opportunities for improving data quality and representativeness, controlling for bias, and allowing humans to oversee and impact computational processes are missed if we do not consider the lifecycle stages upstream from model training and deployment. Yet, very little has been done to date to provide system-level support to data scientists who wish to develop and deploy responsible machine learning methods. We aim to fill this gap and present FairPrep, a design and evaluation framework for fairness-enhancing interventions. FairPrep is based on a developer-centered design, and helps data scientists follow best practices in software engineering and machine learning. As part of our contribution, we identify shortcomings in existing empirical studies for analyzing fairness-enhancing interventions. We then show how FairPrep can be used to measure the impact of sound best practices, such as hyperparameter tuning and feature scaling. In particular, our results suggest that the high variability of the outcomes of fairness-enhancing interventions observed in previous studies is often an artifact of a lack of hyperparameter tuning. Further, we show that the choice of a data cleaning method can impact the effectiveness of fairness-enhancing interventions. |
| Tasks | Decision Making |
| Published | 2019-11-28 |
| URL | https://arxiv.org/abs/1911.12587v1 |
| https://arxiv.org/pdf/1911.12587v1.pdf | |
| PWC | https://paperswithcode.com/paper/fairprep-promoting-data-to-a-first-class |
| Repo | |
| Framework | |
Real-time texturing for 6D object instance detection from RGB Images
| Title | Real-time texturing for 6D object instance detection from RGB Images |
| Authors | Pavel Rojtberg, Arjan Kuijper |
| Abstract | For objected detection, the availability of color cues strongly influences detection rates and is even a prerequisite for many methods. However, when training on synthetic CAD data, this information is not available. We therefore present a method for generating a texture-map from image sequences in real-time. The method relies on 6 degree-of-freedom poses and a 3D-model being available. In contrast to previous works this allows interleaving detection and texturing for upgrading the detector on-the-fly. Our evaluation shows that the acquired texture-map significantly improves detection rates using the LINEMOD detector on RGB images only. Additionally, we use the texture-map to differentiate instances of the same object by surface color. |
| Tasks | |
| Published | 2019-12-13 |
| URL | https://arxiv.org/abs/1912.06404v1 |
| https://arxiv.org/pdf/1912.06404v1.pdf | |
| PWC | https://paperswithcode.com/paper/real-time-texturing-for-6d-object-instance |
| Repo | |
| Framework | |
Toward Automated Website Classification by Deep Learning
| Title | Toward Automated Website Classification by Deep Learning |
| Authors | Fabrizio De Fausti, Francesco Pugliese, Diego Zardetto |
| Abstract | In recent years, the interest in Big Data sources has been steadily growing within the Official Statistic community. The Italian National Institute of Statistics (Istat) is currently carrying out several Big Data pilot studies. One of these studies, the ICT Big Data pilot, aims at exploiting massive amounts of textual data automatically scraped from the websites of Italian enterprises in order to predict a set of target variables (e.g. e-commerce) that are routinely observed by the traditional ICT Survey. In this paper, we show that Deep Learning techniques can successfully address this problem. Essentially, we tackle a text classification task: an algorithm must learn to infer whether an Italian enterprise performs e-commerce from the textual content of its website. To reach this goal, we developed a sophisticated processing pipeline and evaluated its performance through extensive experiments. Our pipeline uses Convolutional Neural Networks and relies on Word Embeddings to encode raw texts into grayscale images (i.e. normalized numeric matrices). Web-scraped texts are huge and have very low signal to noise ratio: to overcome these issues, we adopted a framework known as False Positive Reduction, which has seldom (if ever) been applied before to text classification tasks. Several original contributions enable our processing pipeline to reach good classification results. Empirical evidence shows that our proposal outperforms all the alternative Machine Learning solutions already tested in Istat for the same task. |
| Tasks | Text Classification, Word Embeddings |
| Published | 2019-10-22 |
| URL | https://arxiv.org/abs/1910.09991v1 |
| https://arxiv.org/pdf/1910.09991v1.pdf | |
| PWC | https://paperswithcode.com/paper/toward-automated-website-classification-by |
| Repo | |
| Framework | |
PT-CoDE: Pre-trained Context-Dependent Encoder for Utterance-level Emotion Recognition
| Title | PT-CoDE: Pre-trained Context-Dependent Encoder for Utterance-level Emotion Recognition |
| Authors | Wenxiang Jiao, Michael R. Lyu, Irwin King |
| Abstract | Utterance-level emotion recognition (ULER) is a significant research topic for understanding human behaviors and developing empathetic chatting machines in the artificial intelligence area. Unlike traditional text classification problem, this task is supported by a limited number of datasets, among which most contain inadequate conversations or speeches. Such a data scarcity issue limits the possibility of training larger and more powerful models for this task. Witnessing the success of transfer learning in natural language process (NLP), we propose to pre-train a context-dependent encoder (CoDE) for ULER by learning from unlabeled conversation data. Essentially, CoDE is a hierarchical architecture that contains an utterance encoder and a conversation encoder, making it different from those works that aim to pre-train a universal sentence encoder. Also, we propose a new pre-training task named “conversation completion” (CoCo), which attempts to select the correct answer from candidate answers to fill a masked utterance in a question conversation. The CoCo task is carried out on pure movie subtitles so that our CoDE can be pre-trained in an unsupervised fashion. Finally, the pre-trained CoDE (PT-CoDE) is fine-tuned for ULER and boosts the model performance significantly on five datasets. |
| Tasks | Emotion Recognition, Text Classification, Transfer Learning |
| Published | 2019-10-20 |
| URL | https://arxiv.org/abs/1910.08916v1 |
| https://arxiv.org/pdf/1910.08916v1.pdf | |
| PWC | https://paperswithcode.com/paper/pt-code-pre-trained-context-dependent-encoder |
| Repo | |
| Framework | |
An Adaptive Random Path Selection Approach for Incremental Learning
| Title | An Adaptive Random Path Selection Approach for Incremental Learning |
| Authors | Jathushan Rajasegaran, Munawar Hayat, Salman Khan, Fahad Shahbaz Khan, Ling Shao, Ming-Hsuan Yang |
| Abstract | In a conventional supervised learning setting, a machine learning model has access to examples of all object classes that are desired to be recognized during the inference stage. This results in a fixed model that lacks the flexibility to adapt to new learning tasks. In practical settings, learning tasks often arrive in a sequence and the models must continually learn to increment their previously acquired knowledge. Existing incremental learning approaches fall well below the state-of-the-art cumulative models that use all training classes at once. In this paper, we propose a random path selection algorithm, called Adaptive RPS-Net, that progressively chooses optimal paths for the new tasks while encouraging parameter sharing between tasks. We introduce a new network capacity measure that enables us to automatically switch paths if the already used resources are saturated. Since the proposed path-reuse strategy ensures forward knowledge transfer, our approach is efficient and has considerably less computation overhead. As an added novelty, the proposed model integrates knowledge distillation and retrospection along with the path selection strategy to overcome catastrophic forgetting. In order to maintain an equilibrium between previous and newly acquired knowledge, we propose a simple controller to dynamically balance the model plasticity. Through extensive experiments, we demonstrate that the Adaptive RPS-Net method surpasses the state-of-the-art performance for incremental learning and by utilizing parallel computation this method can run in constant time with nearly the same efficiency as a conventional deep convolutional neural network. |
| Tasks | Transfer Learning |
| Published | 2019-06-03 |
| URL | https://arxiv.org/abs/1906.01120v3 |
| https://arxiv.org/pdf/1906.01120v3.pdf | |
| PWC | https://paperswithcode.com/paper/random-path-selection-for-incremental |
| Repo | |
| Framework | |
Many Faces of Feature Importance: Comparing Built-in and Post-hoc Feature Importance in Text Classification
| Title | Many Faces of Feature Importance: Comparing Built-in and Post-hoc Feature Importance in Text Classification |
| Authors | Vivian Lai, Jon Z. Cai, Chenhao Tan |
| Abstract | Feature importance is commonly used to explain machine predictions. While feature importance can be derived from a machine learning model with a variety of methods, the consistency of feature importance via different methods remains understudied. In this work, we systematically compare feature importance from built-in mechanisms in a model such as attention values and post-hoc methods that approximate model behavior such as LIME. Using text classification as a testbed, we find that 1) no matter which method we use, important features from traditional models such as SVM and XGBoost are more similar with each other, than with deep learning models; 2) post-hoc methods tend to generate more similar important features for two models than built-in methods. We further demonstrate how such similarity varies across instances. Notably, important features do not always resemble each other better when two models agree on the predicted label than when they disagree. |
| Tasks | Feature Importance, Text Classification |
| Published | 2019-10-18 |
| URL | https://arxiv.org/abs/1910.08534v1 |
| https://arxiv.org/pdf/1910.08534v1.pdf | |
| PWC | https://paperswithcode.com/paper/many-faces-of-feature-importance-comparing |
| Repo | |
| Framework | |
Automatic Algorithm Selection In Multi-agent Pathfinding
| Title | Automatic Algorithm Selection In Multi-agent Pathfinding |
| Authors | Devon Sigurdson, Vadim Bulitko, Sven Koenig, Carlos Hernandez, William Yeoh |
| Abstract | In a multi-agent pathfinding (MAPF) problem, agents need to navigate from their start to their goal locations without colliding into each other. There are various MAPF algorithms, including Windowed Hierarchical Cooperative A*, Flow Annotated Replanning, and Bounded Multi-Agent A*. It is often the case that there is no a single algorithm that dominates all MAPF instances. Therefore, in this paper, we investigate the use of deep learning to automatically select the best MAPF algorithm from a portfolio of algorithms for a given MAPF problem instance. Empirical results show that our automatic algorithm selection approach, which uses an off-the-shelf convolutional neural network, is able to outperform any individual MAPF algorithm in our portfolio. |
| Tasks | |
| Published | 2019-06-10 |
| URL | https://arxiv.org/abs/1906.03992v2 |
| https://arxiv.org/pdf/1906.03992v2.pdf | |
| PWC | https://paperswithcode.com/paper/automatic-algorithm-selection-in-multi-agent |
| Repo | |
| Framework | |
Meta Adaptation using Importance Weighted Demonstrations
| Title | Meta Adaptation using Importance Weighted Demonstrations |
| Authors | Kiran Lekkala, Sami Abu-El-Haija, Laurent Itti |
| Abstract | Imitation learning has gained immense popularity because of its high sample-efficiency. However, in real-world scenarios, where the trajectory distribution of most of the tasks dynamically shifts, model fitting on continuously aggregated data alone would be futile. In some cases, the distribution shifts, so much, that it is difficult for an agent to infer the new task. We propose a novel algorithm to generalize on any related task by leveraging prior knowledge on a set of specific tasks, which involves assigning importance weights to each past demonstration. We show experiments where the robot is trained from a diversity of environmental tasks and is also able to adapt to an unseen environment, using few-shot learning. We also developed a prototype robot system to test our approach on the task of visual navigation, and experimental results obtained were able to confirm these suppositions. |
| Tasks | Few-Shot Learning, Imitation Learning, Visual Navigation |
| Published | 2019-11-23 |
| URL | https://arxiv.org/abs/1911.10322v1 |
| https://arxiv.org/pdf/1911.10322v1.pdf | |
| PWC | https://paperswithcode.com/paper/meta-adaptation-using-importance-weighted |
| Repo | |
| Framework | |
QoE-Aware Resource Allocation for Crowdsourced Live Streaming: A Machine Learning Approach
| Title | QoE-Aware Resource Allocation for Crowdsourced Live Streaming: A Machine Learning Approach |
| Authors | Fatima Haouari, Emna Baccour, Aiman Erbad, Amr Mohamed, Mohsen Guizani |
| Abstract | Driven by the tremendous technological advancement of personal devices and the prevalence of wireless mobile network accesses, the world has witnessed an explosion in crowdsourced live streaming. Ensuring a better viewers quality of experience (QoE) is the key to maximize the audiences number and increase streaming providers’ profits. This can be achieved by advocating a geo-distributed cloud infrastructure to allocate the multimedia resources as close as possible to viewers, in order to minimize the access delay and video stalls. Moreover, allocating the exact needed resources beforehand avoids over-provisioning, which may lead to significant costs by the service providers. In the contrary, under-provisioning might cause significant delays to the viewers. In this paper, we introduce a prediction driven resource allocation framework, to maximize the QoE of viewers and minimize the resource allocation cost. First, by exploiting the viewers locations available in our unique dataset, we implement a machine learning model to predict the viewers number near each geo-distributed cloud site. Second, based on the predicted results that showed to be close to the actual values, we formulate an optimization problem to proactively allocate resources at the viewers proximity. Additionally, we will present a trade-off between the video access delay and the cost of resource allocation. |
| Tasks | |
| Published | 2019-06-20 |
| URL | https://arxiv.org/abs/1906.09086v1 |
| https://arxiv.org/pdf/1906.09086v1.pdf | |
| PWC | https://paperswithcode.com/paper/qoe-aware-resource-allocation-for |
| Repo | |
| Framework | |
Evolution of transfer learning in natural language processing
| Title | Evolution of transfer learning in natural language processing |
| Authors | Aditya Malte, Pratik Ratadiya |
| Abstract | In this paper, we present a study of the recent advancements which have helped bring Transfer Learning to NLP through the use of semi-supervised training. We discuss cutting-edge methods and architectures such as BERT, GPT, ELMo, ULMFit among others. Classically, tasks in natural language processing have been performed through rule-based and statistical methodologies. However, owing to the vast nature of natural languages these methods do not generalise well and failed to learn the nuances of language. Thus machine learning algorithms such as Naive Bayes and decision trees coupled with traditional models such as Bag-of-Words and N-grams were used to usurp this problem. Eventually, with the advent of advanced recurrent neural network architectures such as the LSTM, we were able to achieve state-of-the-art performance in several natural language processing tasks such as text classification and machine translation. We talk about how Transfer Learning has brought about the well-known ImageNet moment for NLP. Several advanced architectures such as the Transformer and its variants have allowed practitioners to leverage knowledge gained from unrelated task to drastically fasten convergence and provide better performance on the target task. This survey represents an effort at providing a succinct yet complete understanding of the recent advances in natural language processing using deep learning in with a special focus on detailing transfer learning and its potential advantages. |
| Tasks | Machine Translation, Text Classification, Transfer Learning |
| Published | 2019-10-16 |
| URL | https://arxiv.org/abs/1910.07370v1 |
| https://arxiv.org/pdf/1910.07370v1.pdf | |
| PWC | https://paperswithcode.com/paper/evolution-of-transfer-learning-in-natural |
| Repo | |
| Framework | |
Towards French Smart Building Code: Compliance Checking Based on Semantic Rules
| Title | Towards French Smart Building Code: Compliance Checking Based on Semantic Rules |
| Authors | Nicolas Bus, Ana Roxin, Guillaume Picinbono, Muhammad Fahad |
| Abstract | Manually checking models for compliance against building regulation is a time-consuming task for architects and construction engineers. There is thus a need for algorithms that process information from construction projects and report non-compliant elements. Still automated code-compliance checking raises several obstacles. Building regulations are usually published as human readable texts and their content is often ambiguous or incomplete. Also, the vocabulary used for expressing such regulations is very different from the vocabularies used to express Building Information Models (BIM). Furthermore, the high level of details associated to BIM-contained geometries induces complex calculations. Finally, the level of complexity of the IFC standard also hinders the automation of IFC processing tasks. Model chart, formal rules and pre-processors approach allows translating construction regulations into semantic queries. We further demonstrate the usefulness of this approach through several use cases. We argue our approach is a step forward in bridging the gap between regulation texts and automated checking algorithms. Finally with the recent building ontology BOT recommended by the W3C Linked Building Data Community Group, we identify perspectives for standardizing and extending our approach. |
| Tasks | |
| Published | 2019-10-01 |
| URL | https://arxiv.org/abs/1910.00334v1 |
| https://arxiv.org/pdf/1910.00334v1.pdf | |
| PWC | https://paperswithcode.com/paper/towards-french-smart-building-code-compliance |
| Repo | |
| Framework | |
Machine Translation of Restaurant Reviews: New Corpus for Domain Adaptation and Robustness
| Title | Machine Translation of Restaurant Reviews: New Corpus for Domain Adaptation and Robustness |
| Authors | Alexandre Bérard, Ioan Calapodescu, Marc Dymetman, Claude Roux, Jean-Luc Meunier, Vassilina Nikoulina |
| Abstract | We share a French-English parallel corpus of Foursquare restaurant reviews (https://europe.naverlabs.com/research/natural-language-processing/machine-translation-of-restaurant-reviews), and define a new task to encourage research on Neural Machine Translation robustness and domain adaptation, in a real-world scenario where better-quality MT would be greatly beneficial. We discuss the challenges of such user-generated content, and train good baseline models that build upon the latest techniques for MT robustness. We also perform an extensive evaluation (automatic and human) that shows significant improvements over existing online systems. Finally, we propose task-specific metrics based on sentiment analysis or translation accuracy of domain-specific polysemous words. |
| Tasks | Domain Adaptation, Machine Translation, Sentiment Analysis |
| Published | 2019-10-31 |
| URL | https://arxiv.org/abs/1910.14589v1 |
| https://arxiv.org/pdf/1910.14589v1.pdf | |
| PWC | https://paperswithcode.com/paper/machine-translation-of-restaurant-reviews-new |
| Repo | |
| Framework | |
Fuzzy Knowledge-Based Architecture for Learning and Interaction in Social Robots
| Title | Fuzzy Knowledge-Based Architecture for Learning and Interaction in Social Robots |
| Authors | Mehdi Ghayoumi, Maryam Pourebadi |
| Abstract | In this paper, we introduce an extension of our presented cognitive-based emotion model [27][28]and [30], where we enhance our knowledge-based emotion unit of the architecture by embedding a fuzzy rule-based system to it. The model utilizes the cognitive parameters dependency and their corresponding weights to regulate the robot’s behavior and fuse their behavior data to achieve the final decision in their interaction with the environment. Using this fuzzy system, our previous model can simulate linguistic parameters for better controlling and generating understandable and flexible behaviors in the robots. We implement our model on an assistive healthcare robot, named Robot Nurse Assistant (RNA) and test it with human subjects. Our model records all the emotion states and essential information based on its predefined rules and learning system. Our results show that our robot interacts with patients in a reasonable, faithful way in special conditions which are defined by rules. This work has the potential to provide better on-demand service for clinical experts to monitor the patients’ emotion states and help them make better decisions accordingly. |
| Tasks | |
| Published | 2019-09-14 |
| URL | https://arxiv.org/abs/1909.11004v1 |
| https://arxiv.org/pdf/1909.11004v1.pdf | |
| PWC | https://paperswithcode.com/paper/fuzzy-knowledge-based-architecture-for |
| Repo | |
| Framework | |
CineFilter: Unsupervised Filtering for Real Time Autonomous Camera Systems
| Title | CineFilter: Unsupervised Filtering for Real Time Autonomous Camera Systems |
| Authors | Sudheer Achary, Syed Ashar Javed, Nikita Shravan, K L Bhanu Moorthy, Vineet Gandhi, Anoop Namboodiri |
| Abstract | Learning to mimic the smooth and deliberate camera movement of a human cameraman is an essential requirement for autonomous camera systems. This paper presents a novel formulation for online and real-time estimation of smooth camera trajectories. Many works have focused on global optimization of the trajectory to produce an offline output. Some recent works have tried to extend this to the online setting, but lack either in the quality of the camera trajectories or need large labeled datasets to train their supervised model. We propose two models, one a convex optimization based approach and another a CNN based model, both of which can exploit the temporal trends in the camera behavior. Our model is built in an unsupervised way without any ground truth trajectories and is robust to noisy outliers. We evaluate our models on two different settings namely a basketball dataset and a stage performance dataset and compare against multiple baselines and past approaches. Our models outperform other methods on quantitative and qualitative metrics and produce smooth camera trajectories that are motivated by cinematographic principles. These models can also be easily adopted to run in real-time with a low computational cost, making them fit for a variety of applications. |
| Tasks | |
| Published | 2019-12-11 |
| URL | https://arxiv.org/abs/1912.05636v1 |
| https://arxiv.org/pdf/1912.05636v1.pdf | |
| PWC | https://paperswithcode.com/paper/cinefilter-unsupervised-filtering-for-real |
| Repo | |
| Framework | |
