Lower-level publication types

Recent Submissions 

  1. Zero-Order Robust Nonlinear Model Predictive Control with Ellipsoidal Uncertainty Sets 

    Zanelli, Andrea; Frey, Jonathan; Messerer, Florian; et al. (2021)
    IFAC-PapersOnLine
    In this paper, we propose an efficient zero-order algorithm that can be used to compute an approximate solution to robust optimal control problems (OCP) and robustified nonconvex programs in general. In particular, we focus on robustified OCPs that make use of ellipsoidal uncertainty sets and show that, with the proposed zero-order method, we can efficiently obtain suboptimal, but robustly feasible solutions. The main idea lies in leveraging ...
    Conference Paper
  2. Knowledge Enhanced Machine Learning Pipeline against Diverse Adversarial Attacks 

    Gurel, Nezihe Merve; Qi, Xiangyu; Rimanic, Luka; et al. (2021)
    Proceedings of Machine Learning Research
    Despite the great successes achieved by deep neural networks (DNNs), recent studies show that they are vulnerable against adversarial examples, which aim to mislead DNNs by adding small adversarial perturbations. Several defenses have been proposed against such attacks, while many of them have been adaptively attacked. In this work, we aim to enhance the ML robustness from a different perspective by leveraging domain knowledge: We propose ...
    Conference Paper
  3. Improving Point Cloud Semantic Segmentation by Learning 3D Object Detection 

    Unal, Ozan; Van Gool, Luc; Dai, Dengxin (2021)
    2021 IEEE WINTER CONFERENCE ON APPLICATIONS OF COMPUTER VISION WACV 2021
    Point cloud semantic segmentation plays an essential role in autonomous driving, providing vital information about drivable surfaces and nearby objects that can aid higher level tasks such as path planning and collision avoidance. While current 3D semantic segmentation networks focus on convolutional architectures that perform great for well represented classes, they show a significant drop in performance for underrepresented classes that ...
    Conference Paper
  4. Zero-Pair Image to Image Translation using Domain Conditional Normalization 

    Shukla, Samarth; Romero, Andres; Van Gool, Luc; et al. (2021)
    2021 IEEE WINTER CONFERENCE ON APPLICATIONS OF COMPUTER VISION WACV 2021
    In this paper, we propose an approach based on domain conditional normalization (DCN) for zero-pair image-to-image translation, i.e., translating between two domains which have no paired training data available but each have paired training data with a third domain. We employ a single generator which has an encoder-decoder structure and analyze different implementations of domain conditional normalization to obtain the desired target ...
    Conference Paper
  5. Neural Symbolic Regression that Scales 

    Biggio, Luca; Bendinelli, Tommaso; Neitz, Alexander; et al. (2021)
    Proceedings of Machine Learning Research
    Symbolic equations are at the core of scientific discovery. The task of discovering the underlying equation from a set of input-output pairs is called symbolic regression. Traditionally, symbolic regression methods use hand-designed strategies that do not improve with experience. In this paper, we introduce the first symbolic regression method that leverages large scale pre-training We procedurally generate an unbounded set of equations, ...
    Conference Paper
  6. Scalable Certified Segmentation via Randomized Smoothing 

    Fischer, Marc; Baader, Maximilian; Vechev, Martin (2021)
    Proceedings of Machine Learning Research
    We present a new certification method for image and point cloud segmentation based on randomized smoothing. The method leverages a novel scalable algorithm for prediction and certification that correctly accounts for multiple testing, necessary for ensuring statistical guarantees. The key to our approach is reliance on established multiple-testing correction mechanisms as well as the ability to abstain from classifying single pixels or ...
    Conference Paper
  7. PROGRAML: A Graph-based Program Representation for Data Flow Analysis and Compiler Optimizations 

    Cummins, Chris; Fisches, Zacharias, V; Ben-Nun, Tal; et al. (2021)
    Proceedings of Machine Learning Research
    Machine learning (ML) is increasingly seen as a viable approach for building compiler optimization heuristics, but many ML methods cannot replicate even the simplest of the data flow analyses that are critical to making good optimization decisions. We posit that if ML cannot do that, then it is insufficiently able to reason about programs. We formulate data flow analyses as supervised learning tasks and introduce a large open dataset of ...
    Conference Paper
  8. Towards Visually Explaining Video Understanding Networks with Perturbation 

    Li, Zhenqiang; Wang, Weimin; Li, Zuoyue; et al. (2021)
    2021 IEEE WINTER CONFERENCE ON APPLICATIONS OF COMPUTER VISION (WACV 2021)
    "Making black box models explainable" is a vital problem that accompanies the development of deep learning networks. For networks taking visual information as input, one basic but challenging explanation method is to identify and visualize the input pixels/regions that dominate the network's prediction. However, most existing works focus on explaining networks taking a single image as input and do not consider the temporal relationship ...
    Conference Paper
  9. Bias-Robust Bayesian Optimization via Dueling Bandits 

    Kirschner, Johannes; Krause, Andreas (2021)
    Proceedings of Machine Learning Research
    We consider Bayesian optimization in settings where observations can be adversarially biased, for example by an uncontrolled hidden confounder. Our first contribution is a reduction of the confounded setting to the dueling bandit model. Then we propose a novel approach for dueling bandits based on information-directed sampling (IDS). Thereby, we obtain the first efficient kernelized algorithm for dueling bandits that comes with cumulative ...
    Conference Paper
  10. Facial Emotion Recognition with Noisy Multi-task Annotations 

    Zhang, Siwei; Huang, Zhiwu; Paudel, Danda Pani; et al. (2021)
    2021 IEEE WINTER CONFERENCE ON APPLICATIONS OF COMPUTER VISION (WACV 2021)
    Human emotions can be inferred from facial expressions. However, the annotations of facial expressions are often highly noisy in common emotion coding models, including categorical and dimensional ones. To reduce human labelling effort on multi-task labels, we introduce a new problem of facial emotion recognition with noisy multitask annotations. For this new problem, we suggest a formulation from the point of joint distribution match ...
    Conference Paper
  11. Dynamic Plane Convolutional Occupancy Networks 

    Lionar, Stefan; Emtsev, Daniil; Svilarkovic, Dusan; et al. (2021)
    2021 IEEE WINTER CONFERENCE ON APPLICATIONS OF COMPUTER VISION (WACV 2021)
    Learning-based 3D reconstruction using implicit neural representations has shown promising progress not only at the object level but also in more complicated scenes. In this paper, we propose Dynamic Plane Convolutional Occupancy Networks, a novel implicit representation pushing further the quality of 3D surface reconstruction. The input noisy point clouds are encoded into per-point features that are projected onto multiple 2D dynamic ...
    Conference Paper
  12. TFix: Learning to Fix Coding Errors with a Text-to-Text Transformer 

    Berabi, Berkay; He, Jingxuan; Raychev, Veselin; et al. (2021)
    Proceedings of Machine Learning Research
    The problem of fixing errors in programs has attracted substantial interest over the years. The key challenge for building an effective code fixing tool is to capture a wide range of errors and meanwhile maintain high accuracy. In this paper, we address this challenge and present a new learning-based system, called TFix. TFix works directly on program text and phrases the problem of code fixing as a text-to-text task. In turn, this enables ...
    Conference Paper
  13. Scalable Marginal Likelihood Estimation for Model Selection in Deep Learning 

    Immer, Alexander; Bauer, Matthias; Fortuin, Vincent; et al. (2021)
    Proceedings of Machine Learning Research
    Marginal-likelihood based model-selection, even though promising, is rarely used in deep learning due to estimation difficulties. Instead, most approaches rely on validation data, which may not be readily available. In this work, we present a scalable marginal-likelihood estimation method to select both hyperparameters and network architectures, based on the training data alone. Some hyperparameters can be estimated online during training, ...
    Conference Paper
  14. Uniform Convergence, Adversarial Spheres and a Simple Remedy 

    Bachmann, Gregor; Moosayi-Dezfooli, Seyed-Mohsen; Hofmann, Thomas (2021)
    Proceedings of Machine Learning Research
    Previous work has cast doubt on the general framework of uniform convergence and its ability to explain generalization in neural networks. By considering a specific dataset, it was observed that a neural network completely misclassifies a projection of the training data (adversarial set), rendering any existing generalization bound based on uniform convergence vacuous. We provide an extensive theoretical investigation of the previously ...
    Conference Paper
  15. Consistent regression when oblivious outliers overwhelm 

    d'Orsi, Tommaso; Novikov, Gleb; Steurer, David (2021)
    Proceedings of Machine Learning Research
    We consider a robust linear regression model y = X beta + eta, where an adversary oblivious to the design X epsilon R-nxd may choose eta to corrupt all but an ff fraction of the observations y in an arbitrary way. Prior to our work, even for Gaussian X, no estimator for beta was known to be consistent in this model except for quadratic sample size n greater than or similar to (d/alpha)(2) or for logarithmic inlier fraction alpha >= 1/log ...
    Conference Paper
  16. How rotational invariance of common kernels prevents generalization in high dimensions 

    Donhauser, Konstantin; Wu, Mingqi; Yang, Fanny (2021)
    Proceedings of Machine Learning Research
    Kernel ridge regression is well-known to achieve minimax optimal rates in low-dimensional settings. However, its behavior in high dimensions is much less understood. Recent work establishes consistency for high-dimensional kernel regression for a number of specific assumptions on the data distribution. In this paper, we show that in high dimensions, the rotational invariance property of commonly studied kernels (such as RBF, inner product ...
    Conference Paper
  17. Efficient Performance Bounds for Primal-Dual Reinforcement Learning from Demonstrations 

    Kamoutsi, Angeliki; Banjac, Goran; Lygeros, John (2021)
    Proceedings of Machine Learning Research
    We consider large-scale Markov decision processes with an unknown cost function and address the problem of learning a policy from a finite set of expert demonstrations. We assume that the learner is not allowed to interact with the expert and has no access to reinforcement signal of any kind. Existing inverse reinforcement learning methods come with strong theoretical guarantees, but are computationally expensive, while state-of-the-art ...
    Conference Paper
  18. The Limits of Min-Max Optimization Algorithms: Convergence to Spurious Non-Critical Sets 

    Hsieh, Ya-Ping; Mertikopoulos, Panayotis; Cevher, Volkan (2021)
    Proceedings of Machine Learning Research
    Compared to ordinary function minimization problems, min-max optimization algorithms encounter far greater challenges because of the existence of periodic cycles and similar phenomena. Even though some of these behaviors can be overcome in the convex-concave regime, the general case is considerably more difficult. With this in mind, we take an in-depth look at a comprehensive class of state-of-the art algorithms and prevalent heuristics ...
    Conference Paper
  19. Combining Pessimism with Optimism for Robust and Efficient Model-Based Deep Reinforcement Learning 

    Curi, Sebastian; Bogunovic, Ilija; Krause, Andreas (2021)
    Proceedings of Machine Learning Research
    In real-world tasks, reinforcement learning (RL) agents frequently encounter situations that are not present during training time. To ensure reliable performance, the RL agents need to exhibit robustness against worst-case situations. The robust RL framework addresses this challenge via a worst-case optimization between an agent and an adversary. Previous robust RL algorithms are either sample inefficient, lack robustness guarantees, or ...
    Conference Paper
  20. Stability and performance in MPC using a finite-tail cost 

    Koehler, Johannes; Allgoewer, Frank (2021)
    IFAC-PapersOnLine
    In this paper, we provide a stability and performance analysis of model predictive control (MPC) schemes based on finite-tail costs. We study the MPC formulation originally proposed by Magni et al. (2001) wherein the standard terminal penalty is replaced by a finite-horizon cost of some stabilizing control law. In order to analyse the closed loop, we leverage the more recent technical machinery developed for MPC without terminal ingredients. ...
    Conference Paper

View more