- Conference Paper
Rights / licenseIn Copyright - Non-Commercial Use Permitted
We develop an effective generation of adversarial attacks on neural models that output a sequence of probability distributions rather than a sequence of single values. This setting includes the recently proposed deep probabilistic autoregressive forecasting models that estimate the probability distribution of a time series given its past and achieve state-of-the-art results in a diverse set of application domains. The key technical challenge we address is how to effectively differentiate through the Monte-Carlo estimation of statistics of the output sequence joint distribution. Additionally, we extend prior work on probabilistic forecasting to the Bayesian setting which allows conditioning on future observations, instead of only on past observations. We demonstrate that our approach can successfully generate attacks with small input perturbations in two challenging tasks where robust decision making is crucial – stock market trading and prediction of electricity consumption. Show more
Book titleProceedings of the 37th International Conference on Machine Learning (ICML 2020)
Journal / seriesProceedings of Machine Learning Research
Pages / Article No.
Organisational unit03948 - Vechev, Martin / Vechev, Martin
Related publications and datasets
Is cited by: https://doi.org/10.3929/ethz-b-000498126
NotesConference lecture held on July 14, 2020. Due to the Coronavirus (COVID-19) the conference was conducted virtually.
MoreShow all metadata