TY - JOUR AB - We prove that the observable telegraph signal accompanying the bistability in the photon-blockade-breakdown regime of the driven and lossy Jaynes–Cummings model is the finite-size precursor of what in the thermodynamic limit is a genuine first-order phase transition. We construct a finite-size scaling of the system parameters to a well-defined thermodynamic limit, in which the system remains the same microscopic system, but the telegraph signal becomes macroscopic both in its timescale and intensity. The existence of such a finite-size scaling completes and justifies the classification of the photon-blockade-breakdown effect as a first-order dissipative quantum phase transition. AU - Vukics, A. AU - Dombi, A. AU - Fink, Johannes M AU - Domokos, P. ID - 7451 JF - Quantum SN - 2521-327X TI - Finite-size scaling of the photon-blockade breakdown dissipative quantum phase transition VL - 3 ER - TY - CONF AB - We present a new proximal bundle method for Maximum-A-Posteriori (MAP) inference in structured energy minimization problems. The method optimizes a Lagrangean relaxation of the original energy minimization problem using a multi plane block-coordinate Frank-Wolfe method that takes advantage of the specific structure of the Lagrangean decomposition. We show empirically that our method outperforms state-of-the-art Lagrangean decomposition based algorithms on some challenging Markov Random Field, multi-label discrete tomography and graph matching problems. AU - Swoboda, Paul AU - Kolmogorov, Vladimir ID - 7468 SN - 10636919 T2 - Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition TI - Map inference via block-coordinate Frank-Wolfe algorithm VL - 2019-June ER - TY - CONF AB - Multi-exit architectures, in which a stack of processing layers is interleaved with early output layers, allow the processing of a test example to stop early and thus save computation time and/or energy. In this work, we propose a new training procedure for multi-exit architectures based on the principle of knowledge distillation. The method encourage searly exits to mimic later, more accurate exits, by matching their output probabilities. Experiments on CIFAR100 and ImageNet show that distillation-based training significantly improves the accuracy of early exits while maintaining state-of-the-art accuracy for late ones. The method is particularly beneficial when training data is limited and it allows a straightforward extension to semi-supervised learning,i.e. making use of unlabeled data at training time. Moreover, it takes only afew lines to implement and incurs almost no computational overhead at training time, and none at all at test time. AU - Bui Thi Mai, Phuong AU - Lampert, Christoph ID - 7479 SN - 15505499 T2 - IEEE International Conference on Computer Vision TI - Distillation-based training for multi-exit architectures VL - 2019-October ER - TY - CONF AB - We present a novel class of convolutional neural networks (CNNs) for set functions,i.e., data indexed with the powerset of a finite set. The convolutions are derivedas linear, shift-equivariant functions for various notions of shifts on set functions.The framework is fundamentally different from graph convolutions based on theLaplacian, as it provides not one but several basic shifts, one for each element inthe ground set. Prototypical experiments with several set function classificationtasks on synthetic datasets and on datasets derived from real-world hypergraphsdemonstrate the potential of our new powerset CNNs. AU - Wendler, Chris AU - Alistarh, Dan-Adrian AU - Püschel, Markus ID - 7542 SN - 1049-5258 TI - Powerset convolutional neural networks VL - 32 ER - TY - CONF AB - We propose a new model for detecting visual relationships, such as "person riding motorcycle" or "bottle on table". This task is an important step towards comprehensive structured mage understanding, going beyond detecting individual objects. Our main novelty is a Box Attention mechanism that allows to model pairwise interactions between objects using standard object detection pipelines. The resulting model is conceptually clean, expressive and relies on well-justified training and prediction procedures. Moreover, unlike previously proposed approaches, our model does not introduce any additional complex components or hyperparameters on top of those already required by the underlying detection model. We conduct an experimental evaluation on two datasets, V-COCO and Open Images, demonstrating strong quantitative and qualitative results. AU - Kolesnikov, Alexander AU - Kuznetsova, Alina AU - Lampert, Christoph AU - Ferrari, Vittorio ID - 7640 SN - 9781728150239 T2 - Proceedings of the 2019 International Conference on Computer Vision Workshop TI - Detecting visual relationships using box attention ER -