Added further literature

Signed-off-by: Jim Martens <github@2martens.de>

ma.bib

@@ -44,6 +44,7 @@ general idea: robot drives in environment, autoencoder says it's novel, robot sw
   author      = {Charles Blundell and Julien Cornebise and Koray Kavukcuoglu and Daan Wierstra},
   title       = {Weight Uncertainty in Neural Networks},
   journal     = {arXiv e-prints},
+  year        = {2015},
   date        = {2015-05-21},
   eid         = {arXiv:1505.05424v2},
   eprint      = {1505.05424v2},
@@ -130,6 +131,7 @@ and 17.0% which is considerably better than the previous state-of-the-art. The n
   publisher     = {Curran Associates, Inc.},
   pages         = {6402--6413},
   url           = {http://papers.nips.cc/paper/7219-simple-and-scalable-predictive-uncertainty-estimation-using-deep-ensembles.pdf},
+  __markedentry = {[jim:1]},
   abstract      = {Deep neural networks (NNs) are powerful black box predictors that have recently achieved impressive performance on a wide spectrum of tasks. Quantifying predictive uncertainty in NNs is a challenging and yet unsolved problem. Bayesian NNs, which learn a distribution over weights, are currently the state-of-the-art for estimating predictive uncertainty; however these require significant modifications to the training procedure and are computationally expensive compared to standard (non-Bayesian) NNs. We propose an alternative to Bayesian NNs that is simple to implement, readily parallelizable, requires very little hyperparameter tuning, and yields high quality predictive uncertainty estimates. Through a series of experiments on classification and regression benchmarks, we demonstrate that our method produces well-calibrated uncertainty estimates which are as good or better than approximate Bayesian NNs. To assess robustness to dataset shift, we evaluate the predictive uncertainty on test examples from known and unknown distributions, and show that our method is able to express higher uncertainty on out-of-distribution examples. We demonstrate the scalability of our method by evaluating predictive uncertainty estimates on ImageNet.},
   owner         = {jim},
   timestamp     = {2019.01.02},
@@ -149,4 +151,94 @@ and 17.0% which is considerably better than the previous state-of-the-art. The n
   timestamp = {2019.01.02},
 }
 
+@InProceedings{Ilg2018,
+  author    = {Eddy Ilg and Özgün {\c{C}}i{\c{c}}ek and Silvio Galesso and Aaron Klein and Osama Makansi and Frank Hutter and Thomas Brox},
+  title     = {Uncertainty Estimates and Multi-hypotheses Networks for Optical Flow},
+  booktitle = {Computer Vision {\textendash} {ECCV} 2018},
+  year      = {2018},
+  editor    = {Ferrari, V. and Hebert, M. and Sminchisescu, C. and Weiss, Y.},
+  publisher = {Springer, Cham},
+  pages     = {677--693},
+  doi       = {10.1007/978-3-030-01234-2_40},
+  owner     = {jim},
+  timestamp = {2019.01.03},
+}
+
+@InCollection{Sensoy2018,
+  author    = {Sensoy, Murat and Kaplan, Lance and Kandemir, Melih},
+  title     = {Evidential Deep Learning to Quantify Classification Uncertainty},
+  booktitle = {Advances in Neural Information Processing Systems},
+  year      = {2018},
+  editor    = {S. Bengio and H. Wallach and H. Larochelle and K. Grauman and N. Cesa-Bianchi and R. Garnett},
+  volume    = {31},
+  publisher = {Curran Associates, Inc.},
+  pages     = {3183--3193},
+  url       = {http://papers.nips.cc/paper/7580-evidential-deep-learning-to-quantify-classification-uncertainty.pdf},
+  abstract  = {Deterministic neural nets have been shown to learn effective predictors on a wide range of machine learning problems. However, as the standard approach is to train the network to minimize a prediction loss, the resultant model remains ignorant to its prediction confidence. Orthogonally to Bayesian neural nets that indirectly infer prediction uncertainty through weight uncertainties, we propose explicit modeling of the same using the theory of subjective logic. By placing a Dirichlet distribution on the class probabilities, we treat predictions of a neural net as subjective opinions and learn the function that collects the evidence leading to these opinions by a deterministic neural net from data. The resultant predictor for a multi-class classification problem is another Dirichlet distribution whose parameters are set by the continuous output of a neural net. We provide a preliminary analysis on how the peculiarities of our new loss function drive improved uncertainty estimation. We observe that our method achieves unprecedented success on detection of out-of-distribution queries and endurance against adversarial perturbations.},
+  owner     = {jim},
+  timestamp = {2019.01.03},
+}
+
+@InProceedings{Wu2019,
+  author        = {Anqi Wu and Sebastian Nowozin and Edward Meeds and Richard E. Turner and José Miguel Hernández-Lobato and Alexander L. Gaunt},
+  title         = {Deterministic Variational Inference for Robust Bayesian Neural Networks},
+  booktitle     = {International Conference on Learning Representations},
+  year          = {2019},
+  url           = {https://openreview.net/forum?id=B1l08oAct7},
+  urldate       = {2019-01-03},
+  __markedentry = {[jim:1]},
+  abstract      = {Bayesian neural networks (BNNs) hold great promise as a flexible and principled solution to deal with uncertainty when learning from finite data. Among approaches to realize probabilistic inference in deep neural networks, variational Bayes (VB) is theoretically grounded, generally applicable, and computationally efficient. With wide recognition of potential advantages, why is it that variational Bayes has seen very limited practical use for BNNs in real applications? We argue that variational inference in neural networks is fragile: successful implementations require careful initialization and tuning of prior variances, as well as controlling the variance of Monte Carlo gradient estimates. We provide two innovations that aim to turn VB into a robust inference tool for Bayesian neural networks: first, we introduce a novel deterministic method to approximate moments in neural networks, eliminating gradient variance; second, we introduce a hierarchical prior for parameters and a novel Empirical Bayes procedure for automatically selecting prior variances. Combining these two innovations, the resulting method is highly efficient and robust. On the application of heteroscedastic regression we demonstrate good predictive performance over alternative approaches.},
+  owner         = {jim},
+  timestamp     = {2019.01.03},
+}
+
+@Article{Geifman2018,
+  author        = {Yonatan Geifman and Guy Uziel and Ran El-Yaniv},
+  title         = {Bias-Reduced Uncertainty Estimation for Deep Neural Classifiers},
+  journal       = {arXiv e-prints},
+  date          = {2018-09-30},
+  eid           = {arXiv:1805.08206v3},
+  eprint        = {1805.08206v3},
+  eprintclass   = {cs.LG},
+  eprinttype    = {arXiv},
+  __markedentry = {[jim:1]},
+  abstract      = {We consider the problem of uncertainty estimation in the context of (non-Bayesian) deep neural classification. In this context, all known methods are based on extracting uncertainty signals from a trained network optimized to solve the classification problem at hand. We demonstrate that such techniques tend to introduce biased estimates for instances whose predictions are supposed to be highly confident. We argue that this deficiency is an artifact of the dynamics of training with SGD-like optimizers, and it has some properties similar to overfitting. Based on this observation, we develop an uncertainty estimation algorithm that selectively estimates the uncertainty of highly confident points, using earlier snapshots of the trained model, before their estimates are jittered (and way before they are ready for actual classification). We present extensive experiments indicating that the proposed algorithm provides uncertainty estimates that are consistently better than all known methods.},
+  file          = {:http\://arxiv.org/pdf/1805.08206v3:PDF},
+  keywords      = {cs.LG, stat.ML},
+  owner         = {jim},
+  timestamp     = {2019.01.03},
+}
+
+@Article{Mukhoti2018,
+  author      = {Jishnu Mukhoti and Yarin Gal},
+  title       = {Evaluating Bayesian Deep Learning Methods for Semantic Segmentation},
+  journal     = {arXiv e-prints},
+  date        = {2018-11-30},
+  eid         = {arXiv:1811.12709v1},
+  eprint      = {1811.12709v1},
+  eprintclass = {cs.CV},
+  eprinttype  = {arXiv},
+  abstract    = {Deep learning has been revolutionary for computer vision and semantic segmentation in particular, with Bayesian Deep Learning (BDL) used to obtain uncertainty maps from deep models when predicting semantic classes. This information is critical when using semantic segmentation for autonomous driving for example. Standard semantic segmentation systems have well-established evaluation metrics. However, with BDL's rising popularity in computer vision we require new metrics to evaluate whether a BDL method produces better uncertainty estimates than another method. In this work we propose three such metrics to evaluate BDL models designed specifically for the task of semantic segmentation. We modify DeepLab-v3+, one of the state-of-the-art deep neural networks, and create its Bayesian counterpart using MC dropout and Concrete dropout as inference techniques. We then compare and test these two inference techniques on the well-known Cityscapes dataset using our suggested metrics. Our results provide new benchmarks for researchers to compare and evaluate their improved uncertainty quantification in pursuit of safer semantic segmentation.},
+  file        = {:http\://arxiv.org/pdf/1811.12709v1:PDF},
+  keywords    = {cs.CV},
+  owner       = {jim},
+  timestamp   = {2019.01.03},
+}
+
+@Article{Jankowiak2018,
+  author      = {Martin Jankowiak},
+  title       = {Closed Form Variational Objectives For Bayesian Neural Networks with a Single Hidden Layer},
+  journal     = {arXiv e-prints},
+  date        = {2018-12-02},
+  eid         = {arXiv:1811.00686v2},
+  eprint      = {1811.00686v2},
+  eprintclass = {stat.ML},
+  eprinttype  = {arXiv},
+  abstract    = {In this note we consider setups in which variational objectives for Bayesian neural networks can be computed in closed form. In particular we focus on single-layer networks in which the activation function is piecewise polynomial (e.g. ReLU). In this case we show that for a Normal likelihood and structured Normal variational distributions one can compute a variational lower bound in closed form. In addition we compute the predictive mean and variance in closed form. Finally, we also show how to compute approximate lower bounds for other likelihoods (e.g. softmax classification). In experiments we show how the resulting variational objectives can help improve training and provide fast test time predictions.},
+  file        = {:http\://arxiv.org/pdf/1811.00686v2:PDF},
+  keywords    = {stat.ML, cs.LG},
+  owner       = {jim},
+  timestamp   = {2019.01.03},
+}
+
 @Comment{jabref-meta: databaseType:biblatex;}