| 1. |
|
|
| 2. |
- ODonnell, Michael, et al.
(författare)
-
Registered Replication Report: Dijksterhuis and van Knippenberg (1998)
- 2018
-
Ingår i: Perspectives on Psychological Science. - : SAGE PUBLICATIONS LTD. - 1745-6916 .- 1745-6924. ; 13:2, s. 268-294
-
Tidskriftsartikel (refereegranskat)abstract
- Dijksterhuis and van Knippenberg (1998) reported that participants primed with a category associated with intelligence (professor) subsequently performed 13% better on a trivia test than participants primed with a category associated with a lack of intelligence (soccer hooligans). In two unpublished replications of this study designed to verify the appropriate testing procedures, Dijksterhuis, van Knippenberg, and Holland observed a smaller difference between conditions (2%-3%) as well as a gender difference: Men showed the effect (9.3% and 7.6%), but women did not (0.3% and -0.3%). The procedure used in those replications served as the basis for this multilab Registered Replication Report. A total of 40 laboratories collected data for this project, and 23 of these laboratories met all inclusion criteria. Here we report the meta-analytic results for those 23 direct replications (total N = 4,493), which tested whether performance on a 30-item general-knowledge trivia task differed between these two priming conditions (results of supplementary analyses of the data from all 40 labs, N = 6,454, are also reported). We observed no overall difference in trivia performance between participants primed with the professor category and those primed with the hooligan category (0.14%) and no moderation by gender.
|
|
| 3. |
|
|
| 4. |
- Bendtsen, Katja Maria, et al.
(författare)
-
Particle characterization and toxicity in C57BL/6 mice following instillation of five different diesel exhaust particles designed to differ in physicochemical properties
- 2020
-
Ingår i: Particle and Fibre Toxicology. - : Springer Science and Business Media LLC. - 1743-8977. ; 17:1
-
Tidskriftsartikel (refereegranskat)abstract
- BACKGROUND: Diesel exhaust is carcinogenic and exposure to diesel particles cause health effects. We investigated the toxicity of diesel exhaust particles designed to have varying physicochemical properties in order to attribute health effects to specific particle characteristics. Particles from three fuel types were compared at 13% engine intake O2 concentration: MK1 ultra low sulfur diesel (DEP13) and the two renewable diesel fuels hydrotreated vegetable oil (HVO13) and rapeseed methyl ester (RME13). Additionally, diesel particles from MK1 ultra low sulfur diesel were generated at 9.7% (DEP9.7) and 17% (DEP17) intake O2 concentration. We evaluated physicochemical properties and histopathological, inflammatory and genotoxic responses on day 1, 28, and 90 after single intratracheal instillation in mice compared to reference diesel particles and carbon black. RESULTS: Moderate variations were seen in physical properties for the five particles: primary particle diameter: 15-22 nm, specific surface area: 152-222 m2/g, and count median mobility diameter: 55-103 nm. Larger differences were found in chemical composition: organic carbon/total carbon ratio (0.12-0.60), polycyclic aromatic hydrocarbon content (1-27 μg/mg) and acid-extractable metal content (0.9-16 μg/mg). Intratracheal exposure to all five particles induced similar toxicological responses, with different potency. Lung particle retention was observed in DEP13 and HVO13 exposed mice on day 28 post-exposure, with less retention for the other fuel types. RME exposure induced limited response whereas the remaining particles induced dose-dependent inflammation and acute phase response on day 1. DEP13 induced acute phase response on day 28 and inflammation on day 90. DNA strand break levels were not increased as compared to vehicle, but were increased in lung and liver compared to blank filter extraction control. Neutrophil influx on day 1 correlated best with estimated deposited surface area, but also with elemental carbon, organic carbon and PAHs. DNA strand break levels in lung on day 28 and in liver on day 90 correlated with acellular particle-induced ROS. CONCLUSIONS: We studied diesel exhaust particles designed to differ in physicochemical properties. Our study highlights specific surface area, elemental carbon content, PAHs and ROS-generating potential as physicochemical predictors of diesel particle toxicity.
|
|
| 5. |
- Berg, Axel
(författare)
-
Applications of Diversity and the Self-Attention Mechanism in Neural Networks
- 2022
-
Licentiatavhandling (övrigt vetenskapligt/konstnärligt)abstract
- This thesis covers three contributions in applications of neural networks. The first is related to diversity and ensemble learning, while the other two cover novel applications of the self-attention mechanism. An important aspect of training a neural network is the choice of objective function. Regression via Classification (RvC) is often used to tackle problems in deep learning where the target variable is continuous, but standard regression objectives fail to capture the underlying distance metric of the domain. This can result in better performance of the trained model, but the optimal choice of discrete classes used in RvC is not well understood. In Paper 1, we introduce the concept of label diversity by generalizing the RvC method. By exploiting the fact that labels can be generated in arbitrary ways for continuous and ordinal target variables, we show that using multiple labels can improve the prediction accuracy of a neural network compared to using a single label and provide theoretical justification from ensemble theory. We apply our method to several tasks in computer vision and show increased performance compared to regression and RvC baselines. The performance of a neural network is also influenced by the choice of network architecture, and in the design process it is important to consider the domain of the inputs and its symmetries. Graph neural networks (GNNs) is the family of networks that operates on graphs, where in-formation is propagated between the graph nodes using for example self-attention. However, self-attention can be used for other data domains as well if the inputs can be converted into graphs, which is not always trivial. In Paper 2, we do this for audio by using a complete graph over audio features extracted from different time slots. We apply this technique to the task of keyword spotting and show that a neural network solely based on self-attention is more accurate than previously considered architectures. Finally, in Paper 3 we apply attention-based learning to point cloud processing, where the permutation symmetry must be preserved. In order to make the self-attention mechanism both more efficient and more expressive, we propose a hierarchical approach that allows individual points to interact on both a local and global scale. By extensive experiments on several bench-marks, we show that this approach improves the descriptiveness of the learned features, while simultaneously reducing the computational complexity compared to an architecture that applies self-attention naively on all input points.
|
|
| 6. |
- Berg, Axel, et al.
(författare)
-
Deep ordinal regression with label diversity
- 2021
-
Ingår i: 2020 25th International Conference on Pattern Recognition (ICPR). - 1051-4651. - 9781728188089 ; , s. 2740-2747
-
Konferensbidrag (refereegranskat)abstract
- Regression via classification (RvC) is a common method used for regression problems in deep learning, where the target variable belongs to a set of continuous values. By discretizing the target into a set of non-overlapping classes, it has been shown that training a classifier can improve neural network accuracy compared to using a standard regression approach. However, it is not clear how the set of discrete classes should be chosen and how it affects the overall solution. In this work, we propose that using several discrete data representations simultaneously can improve neural network learning compared to a single representation. Our approach is end-to-end differentiable and can be added as a simple extension to conventional learning methods, such as deep neural networks. We test our method on three challenging tasks and show that our method reduces the prediction error compared to a baseline RvC approach while maintaining a similar model complexity.
|
|
| 7. |
- Berg, Axel, et al.
(författare)
-
Extending GCC-PHAT using Shift Equivariant Neural Networks
- 2022
-
Ingår i: Proceedings of the Annual Conference of the International Speech Communication Association 2022. ; , s. 1791-1795
-
Konferensbidrag (refereegranskat)abstract
- Speaker localization using microphone arrays depends on accurate time delay estimation techniques. For decades, methods based on the generalized cross correlation with phase transform (GCC-PHAT) have been widely adopted for this purpose. Recently, the GCC-PHAT has also been used to provide input features to neural networks in order to remove the effects of noise and reverberation, but at the cost of losing theoretical guarantees in noise-free conditions. We propose a novel approach to extending the GCC-PHAT, where the received signals are filtered using a shift equivariant neural network that preserves the timing information contained in the signals. By extensive experiments we show that our model consistently reduces the error of the GCC-PHAT in adverse environments, with guarantees of exact time delay recovery in ideal conditions.
|
|
| 8. |
- Berg, Axel, et al.
(författare)
-
Keyword Transformer: A Self-Attention Model for Keyword Spotting
- 2021
-
Ingår i: Proc. Interspeech 2021. ; , s. 4249-4253
-
Konferensbidrag (refereegranskat)abstract
- The Transformer architecture has been successful across many domains, including natural language processing, computer vision and speech recognition. In keyword spotting, self-attention has primarily been used on top of convolutional or recurrent encoders. We investigate a range of ways to adapt the Transformer architecture to keyword spotting and introduce the Keyword Transformer (KWT), a fully self-attentional architecture that exceeds state-of-the-art performance across multiple tasks without any pre-training or additional data. Surprisingly, this simple architecture outperforms more complex models that mix convolutional, recurrent and attentive layers. KWT can be used as a drop-in replacement for these models, setting two new benchmark records on the Google Speech Commands dataset with 98.6% and 97.7% accuracy on the 12 and 35-command tasks respectively.
|
|
| 9. |
- Berg, Axel, et al.
(författare)
-
Points to patches: Enabling the use of self-attention for 3D shape recognition
- 2022
-
Ingår i: 2022 26th International Conference on Pattern Recognition (ICPR). - 1051-4651 .- 2831-7475. - 9781665490627 - 9781665490627 ; , s. 528-534
-
Konferensbidrag (refereegranskat)abstract
- While the Transformer architecture has become ubiquitous in the machine learning field, its adaptation to 3D shape recognition is non-trivial. Due to its quadratic computational complexity, the self-attention operator quickly becomes inefficient as the set of input points grows larger. Furthermore, we find that the attention mechanism struggles to find useful connections between individual points on a global scale. In order to alleviate these problems, we propose a two-stage Point Transformer-in-Transformer (Point-TnT) approach which combines local and global attention mechanisms, enabling both individual points and patches of points to attend to each other effectively. Experiments on shape classification show that such an approach provides more useful features for downstream tasks than the baseline Transformer, while also being more computationally efficient. In addition, we also extend our method to feature matching for scene reconstruction, showing that it can be used in conjunction with existing scene reconstruction pipelines.
|
|
| 10. |
- Berg, Axel, et al.
(författare)
-
THE LU SYSTEM FOR DCASE 2024 SOUND EVENT LOCALIZATION AND DETECTION CHALLENGE
- 2024
-
Rapport (övrigt vetenskapligt/konstnärligt)abstract
- This technical report gives an overview of our submission to task 3 of the DCASE 2024 challenge. We present a sound event localization and detection (SELD) system using input features based on trainable neural generalized cross-correlations with phase transform (NGCC-PHAT). With these features together with spectrograms as input to a Transformer-based network, we achieve significant improvements over the baseline method. In addition, we also present an audio-visual version of our system, where distance predictions are updated using depth maps from the panorama video frames.
|
|
