Paper Group ANR 1476
Federated and Differentially Private Learning for Electronic Health Records. Recognizing Arrow Of Time In The Short Stories. Inverting Deep Generative models, One layer at a time. Passport: Enabling Accurate Country-Level Router Geolocation using Inaccurate Sources. Assessing the Local Interpretability of Machine Learning Models. Multi-View Multipl …
Federated and Differentially Private Learning for Electronic Health Records
| Title | Federated and Differentially Private Learning for Electronic Health Records |
| Authors | Stephen R. Pfohl, Andrew M. Dai, Katherine Heller |
| Abstract | The use of collaborative and decentralized machine learning techniques such as federated learning have the potential to enable the development and deployment of clinical risk predictions models in low-resource settings without requiring sensitive data be shared or stored in a central repository. This process necessitates communication of model weights or updates between collaborating entities, but it is unclear to what extent patient privacy is compromised as a result. To gain insight into this question, we study the efficacy of centralized versus federated learning in both private and non-private settings. The clinical prediction tasks we consider are the prediction of prolonged length of stay and in-hospital mortality across thirty one hospitals in the eICU Collaborative Research Database. We find that while it is straightforward to apply differentially private stochastic gradient descent to achieve strong privacy bounds when training in a centralized setting, it is considerably more difficult to do so in the federated setting. |
| Tasks | |
| Published | 2019-11-13 |
| URL | https://arxiv.org/abs/1911.05861v1 |
| https://arxiv.org/pdf/1911.05861v1.pdf | |
| PWC | https://paperswithcode.com/paper/federated-and-differentially-private-learning |
| Repo | |
| Framework | |
Recognizing Arrow Of Time In The Short Stories
| Title | Recognizing Arrow Of Time In The Short Stories |
| Authors | Fahimeh Hosseini, Hosein Fooladi, Mohammad Reza Samsami |
| Abstract | Recognizing arrow of time in short stories is a challenging task. i.e., given only two paragraphs, determining which comes first and which comes next is a difficult task even for humans. In this paper, we have collected and curated a novel dataset for tackling this challenging task. We have shown that a pre-trained BERT architecture achieves reasonable accuracy on the task, and outperforms RNN-based architectures. |
| Tasks | |
| Published | 2019-03-25 |
| URL | http://arxiv.org/abs/1903.10548v1 |
| http://arxiv.org/pdf/1903.10548v1.pdf | |
| PWC | https://paperswithcode.com/paper/recognizing-arrow-of-time-in-the-short |
| Repo | |
| Framework | |
Inverting Deep Generative models, One layer at a time
| Title | Inverting Deep Generative models, One layer at a time |
| Authors | Qi Lei, Ajil Jalal, Inderjit S. Dhillon, Alexandros G. Dimakis |
| Abstract | We study the problem of inverting a deep generative model with ReLU activations. Inversion corresponds to finding a latent code vector that explains observed measurements as much as possible. In most prior works this is performed by attempting to solve a non-convex optimization problem involving the generator. In this paper we obtain several novel theoretical results for the inversion problem. We show that for the realizable case, single layer inversion can be performed exactly in polynomial time, by solving a linear program. Further, we show that for multiple layers, inversion is NP-hard and the pre-image set can be non-convex. For generative models of arbitrary depth, we show that exact recovery is possible in polynomial time with high probability, if the layers are expanding and the weights are randomly selected. Very recent work analyzed the same problem for gradient descent inversion. Their analysis requires significantly higher expansion (logarithmic in the latent dimension) while our proposed algorithm can provably reconstruct even with constant factor expansion. We also provide provable error bounds for different norms for reconstructing noisy observations. Our empirical validation demonstrates that we obtain better reconstructions when the latent dimension is large. |
| Tasks | |
| Published | 2019-06-18 |
| URL | https://arxiv.org/abs/1906.07437v2 |
| https://arxiv.org/pdf/1906.07437v2.pdf | |
| PWC | https://paperswithcode.com/paper/inverting-deep-generative-models-one-layer-at |
| Repo | |
| Framework | |
Passport: Enabling Accurate Country-Level Router Geolocation using Inaccurate Sources
| Title | Passport: Enabling Accurate Country-Level Router Geolocation using Inaccurate Sources |
| Authors | Muzammil Abdul Rehman, Sharon Goldberg, David Choffnes |
| Abstract | When does Internet traffic cross international borders? This question has major geopolitical, legal and social implications and is surprisingly difficult to answer. A critical stumbling block is a dearth of tools that accurately map routers traversed by Internet traffic to the countries in which they are located. This paper presents Passport: a new approach for efficient, accurate country-level router geolocation and a system that implements it. Passport provides location predictions with limited active measurements, using machine learning to combine information from IP geolocation databases, router hostnames, whois records, and ping measurements. We show that Passport substantially outperforms existing techniques, and identify cases where paths traverse countries with implications for security, privacy, and performance. |
| Tasks | |
| Published | 2019-05-12 |
| URL | https://arxiv.org/abs/1905.04651v3 |
| https://arxiv.org/pdf/1905.04651v3.pdf | |
| PWC | https://paperswithcode.com/paper/passport-enabling-accurate-country-level |
| Repo | |
| Framework | |
Assessing the Local Interpretability of Machine Learning Models
| Title | Assessing the Local Interpretability of Machine Learning Models |
| Authors | Dylan Slack, Sorelle A. Friedler, Carlos Scheidegger, Chitradeep Dutta Roy |
| Abstract | The increasing adoption of machine learning tools has led to calls for accountability via model interpretability. But what does it mean for a machine learning model to be interpretable by humans, and how can this be assessed? We focus on two definitions of interpretability that have been introduced in the machine learning literature: simulatability (a user’s ability to run a model on a given input) and “what if” local explainability (a user’s ability to correctly determine a model’s prediction under local changes to the input, given knowledge of the model’s original prediction). Through a user study with 1,000 participants, we test whether humans perform well on tasks that mimic the definitions of simulatability and “what if” local explainability on models that are typically considered locally interpretable. To track the relative interpretability of models, we employ a simple metric, the runtime operation count on the simulatability task. We find evidence that as the number of operations increases, participant accuracy on the local interpretability tasks decreases. In addition, this evidence is consistent with the common intuition that decision trees and logistic regression models are interpretable and are more interpretable than neural networks. |
| Tasks | |
| Published | 2019-02-09 |
| URL | https://arxiv.org/abs/1902.03501v2 |
| https://arxiv.org/pdf/1902.03501v2.pdf | |
| PWC | https://paperswithcode.com/paper/assessing-the-local-interpretability-of |
| Repo | |
| Framework | |
Multi-View Multiple Clusterings using Deep Matrix Factorization
| Title | Multi-View Multiple Clusterings using Deep Matrix Factorization |
| Authors | Shaowei Wei, Jun Wang, Guoxian Yu, Carlotta, Xiangliang Zhang |
| Abstract | Multi-view clustering aims at integrating complementary information from multiple heterogeneous views to improve clustering results. Existing multi-view clustering solutions can only output a single clustering of the data. Due to their multiplicity, multi-view data, can have different groupings that are reasonable and interesting from different perspectives. However, how to find multiple, meaningful, and diverse clustering results from multi-view data is still a rarely studied and challenging topic in multi-view clustering and multiple clusterings. In this paper, we introduce a deep matrix factorization based solution (DMClusts) to discover multiple clusterings. DMClusts gradually factorizes multi-view data matrices into representational subspaces layer-by-layer and generates one clustering in each layer. To enforce the diversity between generated clusterings, it minimizes a new redundancy quantification term derived from the proximity between samples in these subspaces. We further introduce an iterative optimization procedure to simultaneously seek multiple clusterings with quality and diversity. Experimental results on benchmark datasets confirm that DMClusts outperforms state-of-the-art multiple clustering solutions. |
| Tasks | |
| Published | 2019-11-26 |
| URL | https://arxiv.org/abs/1911.11396v1 |
| https://arxiv.org/pdf/1911.11396v1.pdf | |
| PWC | https://paperswithcode.com/paper/multi-view-multiple-clusterings-using-deep |
| Repo | |
| Framework | |
Continuous-Variable Quantum Key Distribution with a Real Local Oscillator and without Auxiliary Signals
| Title | Continuous-Variable Quantum Key Distribution with a Real Local Oscillator and without Auxiliary Signals |
| Authors | Sebastian Kleis, Max Rueckmann, Christian G. Schaeffer |
| Abstract | Continuous-variable quantum key distribution (CV-QKD) is realized with coherent detection and is therefore very suitable for a cost-efficient implementation. The major challenge in CV-QKD is mitigation of laser phase noise at a signal to noise ratio of much less than 0 dB. So far, this has been achieved with a remote local oscillator or with auxiliary signals. For the first time, we experimentally demonstrate that CV-QKD can be performed with a real local oscillator and without auxiliary signals which is achieved by applying Machine Learning methods. It is shown that, with the most established discrete modulation protocol, the experimental system works down to a quantum channel signal to noise ratio of -19.1 dB. The performance of the experimental system allows CV-QKD at a key rate of 9.2 Mbit/s over a fiber distance of 26 km. After remote local oscillator and auxiliary signal aided CV-QKD, this could mark a starting point for a third generation of CV-QKD systems that are even more attractive for a wide implementation because they are almost identical to standard coherent systems. |
| Tasks | |
| Published | 2019-08-02 |
| URL | https://arxiv.org/abs/1908.03625v1 |
| https://arxiv.org/pdf/1908.03625v1.pdf | |
| PWC | https://paperswithcode.com/paper/continuous-variable-quantum-key-distribution |
| Repo | |
| Framework | |
On the Realization and Analysis of Circular Harmonic Transforms for Feature Detection
| Title | On the Realization and Analysis of Circular Harmonic Transforms for Feature Detection |
| Authors | Hugh L Kennedy |
| Abstract | Circular-harmonic spectra are a compact representation of local image features in two dimensions. It is well known that the computational complexity of such transforms is greatly reduced when polar separability is exploited in steerable filter-banks. Further simplifications are possible when Cartesian separability is incorporated using the radial apodization (i.e. weight, window, or taper) described here, as a consequence of the Laguerre/Hermite correspondence over polar/Cartesian coordinates. The chosen form also mitigates undesirable discretization artefacts due to angular aliasing. The possible utility of circular-harmonic spectra for the description of simple features is illustrated using real data from an airborne electro-optic sensor. The spectrum is deployed in a test-statistic to detect and characterize corners of arbitrary angle and orientation (i.e. wedges). The test-statistic considers uncertainty due to finite sampling and clutter/noise. |
| Tasks | |
| Published | 2019-07-29 |
| URL | https://arxiv.org/abs/1907.12165v4 |
| https://arxiv.org/pdf/1907.12165v4.pdf | |
| PWC | https://paperswithcode.com/paper/on-the-realization-and-analysis-of-circular |
| Repo | |
| Framework | |
Deep Reinforcement Learning For Modeling Chit-Chat Dialog With Discrete Attributes
| Title | Deep Reinforcement Learning For Modeling Chit-Chat Dialog With Discrete Attributes |
| Authors | Chinnadhurai Sankar, Sujith Ravi |
| Abstract | Open domain dialog systems face the challenge of being repetitive and producing generic responses. In this paper, we demonstrate that by conditioning the response generation on interpretable discrete dialog attributes and composed attributes, it helps improve the model perplexity and results in diverse and interesting non-redundant responses. We propose to formulate the dialog attribute prediction as a reinforcement learning (RL) problem and use policy gradients methods to optimize utterance generation using long-term rewards. Unlike existing RL approaches which formulate the token prediction as a policy, our method reduces the complexity of the policy optimization by limiting the action space to dialog attributes, thereby making the policy optimization more practical and sample efficient. We demonstrate this with experimental and human evaluations. |
| Tasks | |
| Published | 2019-07-05 |
| URL | https://arxiv.org/abs/1907.02848v2 |
| https://arxiv.org/pdf/1907.02848v2.pdf | |
| PWC | https://paperswithcode.com/paper/deep-reinforcement-learning-for-modeling-chit |
| Repo | |
| Framework | |
A Two-Step Graph Convolutional Decoder for Molecule Generation
| Title | A Two-Step Graph Convolutional Decoder for Molecule Generation |
| Authors | Xavier Bresson, Thomas Laurent |
| Abstract | We propose a simple auto-encoder framework for molecule generation. The molecular graph is first encoded into a continuous latent representation $z$, which is then decoded back to a molecule. The encoding process is easy, but the decoding process remains challenging. In this work, we introduce a simple two-step decoding process. In a first step, a fully connected neural network uses the latent vector $z$ to produce a molecular formula, for example CO$_2$ (one carbon and two oxygen atoms). In a second step, a graph convolutional neural network uses the same latent vector $z$ to place bonds between the atoms that were produced in the first step (for example a double bond will be placed between the carbon and each of the oxygens). This two-step process, in which a bag of atoms is first generated, and then assembled, provides a simple framework that allows us to develop an efficient molecule auto-encoder. Numerical experiments on basic tasks such as novelty, uniqueness, validity and optimized chemical property for the 250k ZINC molecules demonstrate the performances of the proposed system. Particularly, we achieve the highest reconstruction rate of 90.5%, improving the previous rate of 76.7%. We also report the best property improvement results when optimization is constrained by the molecular distance between the original and generated molecules. |
| Tasks | |
| Published | 2019-06-08 |
| URL | https://arxiv.org/abs/1906.03412v2 |
| https://arxiv.org/pdf/1906.03412v2.pdf | |
| PWC | https://paperswithcode.com/paper/a-two-step-graph-convolutional-decoder-for |
| Repo | |
| Framework | |
A Data Driven Method of Optimizing Feedforward Compensator for Autonomous Vehicle
| Title | A Data Driven Method of Optimizing Feedforward Compensator for Autonomous Vehicle |
| Authors | Tianyu Shi, Pin Wang, Ching-Yao Chan, Chonghao Zou |
| Abstract | A reliable controller is critical and essential for the execution of safe and smooth maneuvers of an autonomous vehicle.The controller must be robust to external disturbances, such as road surface, weather, and wind conditions, and so on.It also needs to deal with the internal parametric variations of vehicle sub-systems, including power-train efficiency, measurement errors, time delay,so on.Moreover, as in most production vehicles, the low-control commands for the engine, brake, and steering systems are delivered through separate electronic control units.These aforementioned factors introduce opaque and ineffectiveness issues in controller performance.In this paper, we design a feed-forward compensate process via a data-driven method to model and further optimize the controller performance.We apply the principal component analysis to the extraction of most influential features.Subsequently,we adopt a time delay neural network and include the accuracy of the predicted error in a future time horizon.Utilizing the predicted error,we then design a feed-forward compensate process to improve the control performance.Finally,we demonstrate the effectiveness of the proposed feed-forward compensate process in simulation scenarios. |
| Tasks | |
| Published | 2019-01-31 |
| URL | http://arxiv.org/abs/1901.11212v2 |
| http://arxiv.org/pdf/1901.11212v2.pdf | |
| PWC | https://paperswithcode.com/paper/a-data-driven-method-of-optimizing |
| Repo | |
| Framework | |
The Future of Misinformation Detection: New Perspectives and Trends
| Title | The Future of Misinformation Detection: New Perspectives and Trends |
| Authors | Bin Guo, Yasan Ding, Lina Yao, Yunji Liang, Zhiwen Yu |
| Abstract | The massive spread of misinformation in social networks has become a global risk, implicitly influencing public opinion and threatening social/political development. Misinformation detection (MID) has thus become a surging research topic in recent years. As a promising and rapid developing research field, we find that many efforts have been paid to new research problems and approaches of MID. Therefore, it is necessary to give a comprehensive review of the new research trends of MID. We first give a brief review of the literature history of MID, based on which we present several new research challenges and techniques of it, including early detection, detection by multimodal data fusion, and explanatory detection. We further investigate the extraction and usage of various crowd intelligence in MID, which paves a promising way to tackle MID challenges. Finally, we give our own views on the open issues and future research directions of MID, such as model adaptivity/generality to new events, embracing of novel machine learning models, explanatory detection models, and so on. |
| Tasks | |
| Published | 2019-09-09 |
| URL | https://arxiv.org/abs/1909.03654v1 |
| https://arxiv.org/pdf/1909.03654v1.pdf | |
| PWC | https://paperswithcode.com/paper/the-future-of-misinformation-detection-new |
| Repo | |
| Framework | |
Statistical and Computational Trade-Offs in Kernel K-Means
| Title | Statistical and Computational Trade-Offs in Kernel K-Means |
| Authors | Daniele Calandriello, Lorenzo Rosasco |
| Abstract | We investigate the efficiency of k-means in terms of both statistical and computational requirements. More precisely, we study a Nystr"om approach to kernel k-means. We analyze the statistical properties of the proposed method and show that it achieves the same accuracy of exact kernel k-means with only a fraction of computations. Indeed, we prove under basic assumptions that sampling $\sqrt{n}$ Nystr"om landmarks allows to greatly reduce computational costs without incurring in any loss of accuracy. To the best of our knowledge this is the first result of this kind for unsupervised learning. |
| Tasks | |
| Published | 2019-08-27 |
| URL | https://arxiv.org/abs/1908.10284v1 |
| https://arxiv.org/pdf/1908.10284v1.pdf | |
| PWC | https://paperswithcode.com/paper/statistical-and-computational-trade-offs-in-2 |
| Repo | |
| Framework | |
Alleviating Sequence Information Loss with Data Overlapping and Prime Batch Sizes
| Title | Alleviating Sequence Information Loss with Data Overlapping and Prime Batch Sizes |
| Authors | Noémien Kocher, Christian Scuito, Lorenzo Tarantino, Alexandros Lazaridis, Andreas Fischer, Claudiu Musat |
| Abstract | In sequence modeling tasks the token order matters, but this information can be partially lost due to the discretization of the sequence into data points. In this paper, we study the imbalance between the way certain token pairs are included in data points and others are not. We denote this a token order imbalance (TOI) and we link the partial sequence information loss to a diminished performance of the system as a whole, both in text and speech processing tasks. We then provide a mechanism to leverage the full token order information -Alleviated TOI- by iteratively overlapping the token composition of data points. For recurrent networks, we use prime numbers for the batch size to avoid redundancies when building batches from overlapped data points. The proposed method achieved state of the art performance in both text and speech related tasks. |
| Tasks | |
| Published | 2019-09-18 |
| URL | https://arxiv.org/abs/1909.08700v1 |
| https://arxiv.org/pdf/1909.08700v1.pdf | |
| PWC | https://paperswithcode.com/paper/alleviating-sequence-information-loss-with |
| Repo | |
| Framework | |
Adaptive Robot-Assisted Feeding: An Online Learning Framework for Acquiring Previously-Unseen Food Items
| Title | Adaptive Robot-Assisted Feeding: An Online Learning Framework for Acquiring Previously-Unseen Food Items |
| Authors | Ethan K. Gordon, Xiang Meng, Matt Barnes, Tapomayukh Bhattacharjee, Siddhartha S. Srinivasa |
| Abstract | A successful robot-assisted feeding system requires bite acquisition of a wide variety of food items. It needs to adapt to changing user food preferences under uncertain visual and physical environments. Different food items in different environmental conditions may require different manipulation strategies for successful bite acquisition. Therefore, a key challenge is to handle previously-unseen food items with very different success rate distributions over strategy. Combining low-level controllers and planners into discrete action trajectories, we show that the problem can be represented using a linear contextual bandit setting. We construct a simulated environment using a doubly-robust loss estimate from previously-seen food items, which we use to tune the parameters of off-the-shelf contextual bandit algorithms. Finally, we demonstrate empirically on a robot-assisted feeding system that, even starting with a model trained on thousands of skewering attempts on dissimilar previously-seen food items, $\epsilon$-greedy and LinUCB algorithms can quickly converge to the most successful manipulation strategy. |
| Tasks | |
| Published | 2019-08-19 |
| URL | https://arxiv.org/abs/1908.07088v3 |
| https://arxiv.org/pdf/1908.07088v3.pdf | |
| PWC | https://paperswithcode.com/paper/learning-from-failures-in-robot-assisted |
| Repo | |
| Framework | |
