Share This Author
DeepDB: Learn from Data, not from Queries!
- Benjamin Hilprecht, Andreas Schmidt, Moritz Kulessa, Alejandro Molina, K. Kersting, Carsten Binnig
- Computer ScienceProc. VLDB Endow.
- 2 September 2019
The results of the empirical evaluation demonstrate that the data-driven approach not only provides better accuracy than state-of-the-art learned components but also generalizes better to unseen queries.
Padé Activation Units: End-to-end Learning of Flexible Activation Functions in Deep Networks
This work demonstrates how to eliminate the reliance on first picking fixed activation functions by using flexible parametric rational functions instead, and the resulting Pade Activation Units (PAUs) can both approximate common activation functions and also learn new ones while providing compact representations.
Mixed Sum-Product Networks: A Deep Architecture for Hybrid Domains
- Alejandro Molina, Antonio Vergari, Nicola Di Mauro, Sriraam Natarajan, F. Esposito, K. Kersting
- Computer ScienceAAAI
- 29 April 2018
This work proposes the first trainable probabilistic deep architecture for hybrid domains that features tractable queries and relieves the user from deciding a-priori the parametric form of the random variables but is still expressive enough to effectively approximate any distribution and permits efficient learning and inference.
Random Sum-Product Networks: A Simple and Effective Approach to Probabilistic Deep Learning
This work follows a simple “deep learning” approach, by generating unspecialized random structures, scalable to millions of parameters, and subsequently applying GPUbased optimization, which yields well-calibrated uncertainties and stands out among most deep generative and discriminative models in being robust to missing features and being able to detect anomalies.
SPFlow: An Easy and Extensible Library for Deep Probabilistic Learning using Sum-Product Networks
We introduce SPFlow, an open-source Python library providing a simple interface to inference, learning and manipulation routines for deep and tractable probabilistic models called Sum-Product…
Model-based Approximate Query Processing
- Moritz Kulessa, Alejandro Molina, Carsten Binnig, Benjamin Hilprecht, K. Kersting
- Computer ScienceArXiv
- 15 November 2018
A new approach to AQP is presented called Model-based AQP that leverages generative models learned over the complete database to answer SQL queries at interactive speeds and can in many scenarios return more accurate results in a shorter runtime.
Sum-Product Autoencoding: Encoding and Decoding Representations Using Sum-Product Networks
- Antonio Vergari, Robert Peharz, Nicola Di Mauro, Alejandro Molina, K. Kersting, F. Esposito
- Computer ScienceAAAI
- 29 April 2018
The experimental results on several multi-label classification problems demonstrate that SPAE is competitive with state-of-the-art autoencoder architectures, even if the SPNs were never trained to reconstruct their inputs.
Probabilistic Deep Learning using Random Sum-Product Networks
This paper makes a drastic simplification and uses random SPN structures which are trained in a "classical deep learning manner", i.e. employing automatic differentiation, SGD, and GPU support, and yields prediction results comparable to deep neural networks, while still being interpretable as generative model and maintaining well-calibrated uncertainties.
Perils of Zero-Interaction Security in the Internet of Things
- Mikhail Fomichev, Max Maass, Lars Almon, Alejandro Molina, M. Hollick
- Computer ScienceProc. ACM Interact. Mob. Wearable Ubiquitous…
- 22 January 2019
This paper collects and releases the most comprehensive dataset in the domain to date, containing over 4250 hours of audio recordings and 1 billion sensor readings from three different scenarios, and evaluates five state-of-the-art schemes based on these data.
Einsum Networks: Fast and Scalable Learning of Tractable Probabilistic Circuits
This paper proposes EiNets, a novel implementation design for PCs that combines a large number of arithmetic operations in a single monolithic einsum-operation, leading to speedups and memory savings of up to two orders of magnitude, in comparison to previous implementations.