Author pages are created from data sourced from our academic publisher partnerships and public sources.
Share This Author
Convolutional Recurrent Neural Networks for Polyphonic Sound Event Detection
- Emre Çakir, Giambattista Parascandolo, T. Heittola, H. Huttunen, T. Virtanen
- Computer ScienceIEEE/ACM Transactions on Audio, Speech, and…
- 9 December 2015
This work combines these two approaches in a convolutional recurrent neural network (CRNN) and applies it on a polyphonic sound event detection task and observes a considerable improvement for four different datasets consisting of everyday sound events.
Avoiding Discrimination through Causal Reasoning
- Niki Kilbertus, Mateo Rojas-Carulla, Giambattista Parascandolo, Moritz Hardt, D. Janzing, B. Schölkopf
- Computer ScienceNIPS
- 8 June 2017
This work crisply articulate why and when observational criteria fail, thus formalizing what was before a matter of opinion and put forward natural causal non-discrimination criteria and develop algorithms that satisfy them.
Recurrent neural networks for polyphonic sound event detection in real life recordings
- Giambattista Parascandolo, H. Huttunen, T. Virtanen
- Computer ScienceIEEE International Conference on Acoustics…
- 20 March 2016
In this paper we present an approach to polyphonic sound event detection in real life recordings based on bi-directional long short term memory (BLSTM) recurrent neural networks (RNNs). A single…
Sound Event Detection in Multichannel Audio Using Spatial and Harmonic Features
- Sharath Adavanne, Giambattista Parascandolo, Pasi Pertilä, T. Heittola, T. Virtanen
- Computer Science, PhysicsDCASE
- 7 June 2017
The proposed SED system is compared against the state of the art mono channel method on the development subset of TUT sound events detection 2016 database and the usage of spatial and harmonic features are shown to improve the performance of SED.
Learning explanations that are hard to vary
- Giambattista Parascandolo, Alexander Neitz, Antonio Orvieto, Luigi Gresele, B. Schölkopf
- Computer ScienceICLR
- 1 September 2020
It is shown that averaging gradients across examples -- akin to a logical OR of patterns -- can favor memorization and `patchwork' solutions that sew together different strategies, instead of identifying invariances.
DCASE 2016 Acoustic Scene Classification Using Convolutional Neural Networks
- M. Valenti, Aleksandr Diment, Giambattista Parascandolo, S. Squartini, T. Virtanen
- Computer ScienceDCASE
This workshop paper presents the use of a convolutional neural network trained to classify short sequences of audio, represented by their log-mel spectrogram, and proposes a training method that can be used when the system validation performance saturates as the training proceeds.
Learning Independent Causal Mechanisms
- Giambattista Parascandolo, Mateo Rojas-Carulla, Niki Kilbertus, B. Schölkopf
- Computer ScienceICML
- 4 December 2017
This work develops an algorithm to recover a set of independent (inverse) mechanisms from a sets of transformed data points, based on aset of experts that compete for data generated by the mechanisms, driving specialization.
Convolutional recurrent neural networks for bird audio detection
- Emre Çakir, Sharath Adavanne, Giambattista Parascandolo, K. Drossos, T. Virtanen
- Computer Science25th European Signal Processing Conference…
- 7 March 2017
In the proposed method, convolutional layers extract high dimensional, local frequency shift invariant features, while recurrent layers capture longer term dependencies between the features extracted from short time frames.
Taming the waves: sine as activation function in deep neural networks
This paper formally characterize why deep neural networks can indeed often be difficult to train even in very simple scenarios, and describes how the presence of infinitely many and shallow local minima emerges from the architecture.
Neural Symbolic Regression that Scales
- L. Biggio, Tommaso Bendinelli, Alexander Neitz, Aurélien Lucchi, Giambattista Parascandolo
- Computer Science, MathematicsICML
- 11 June 2021
This paper procedurally generate an unbounded set of equations, and simultaneously pre-train a Transformer to predict the symbolic equation from a corresponding set of input-output-pairs, and shows empirically that this approach can re-discover a set of well-known physical equations and that it improves over time with more data and compute.