Summary
"Deep Learning (DL) has reached unparalleled performance in many domains. However, this impressive performance typically comes at the cost of gathering large datasets and training massive models, requiring extended time and prohibitive costs. Significant research efforts are being invested in improving DL training efficiency, i.e., the amount of time, data, and resources required to train these models, by changing the model (e.g., architecture, numerical precision) or the training algorithm (e.g., parallelization). Other modifications aim to address critical issues, such as credibility and over-confidence, which hinder the implementation of DL in the real world. However, such modifications often cause an unexplained degradation in the generalization performance of DL to unseen data. Recent findings suggest that this degradation is caused by changes to the hidden algorithmic bias of the training algorithm and model. This bias selects a specific solution from all solutions which fit the data. After years of trial-and-error, this bias in DL is often at a ""sweet spot"" which implicitly allows ANNs to learn well, due to unknown key design choices. But performance typically degrades when these choices change. Therefore, understanding and controlling algorithmic bias is the key to unlocking the true potential of deep learning.
Our goal is to develop a rigorous theory of algorithmic bias in DL and to apply it to alleviate critical practical bottlenecks that prevent such models from scaling up or implemented in real-world applications.
Our approach has three objectives: (1) identify the algorithmic biases affecting DL; (2) understand how these biases affect the functional capabilities and generalization performance; and (3) control these biases to alleviate critical practical bottlenecks. To demonstrate the feasibility of this challenging project, we describe how recent advances and concrete preliminary results enable us to effectively approach all these objectives."
Our goal is to develop a rigorous theory of algorithmic bias in DL and to apply it to alleviate critical practical bottlenecks that prevent such models from scaling up or implemented in real-world applications.
Our approach has three objectives: (1) identify the algorithmic biases affecting DL; (2) understand how these biases affect the functional capabilities and generalization performance; and (3) control these biases to alleviate critical practical bottlenecks. To demonstrate the feasibility of this challenging project, we describe how recent advances and concrete preliminary results enable us to effectively approach all these objectives."
Unfold all
/
Fold all
More information & hyperlinks
| Web resources: | https://cordis.europa.eu/project/id/101039436 |
| Start date: | 01-06-2022 |
| End date: | 31-05-2027 |
| Total budget - Public funding: | 1 500 000,00 Euro - 1 500 000,00 Euro |
Cordis data
Original description
"Deep Learning (DL) has reached unparalleled performance in many domains. However, this impressive performance typically comes at the cost of gathering large datasets and training massive models, requiring extended time and prohibitive costs. Significant research efforts are being invested in improving DL training efficiency, i.e., the amount of time, data, and resources required to train these models, by changing the model (e.g., architecture, numerical precision) or the training algorithm (e.g., parallelization). Other modifications aim to address critical issues, such as credibility and over-confidence, which hinder the implementation of DL in the real world. However, such modifications often cause an unexplained degradation in the generalization performance of DL to unseen data. Recent findings suggest that this degradation is caused by changes to the hidden algorithmic bias of the training algorithm and model. This bias selects a specific solution from all solutions which fit the data. After years of trial-and-error, this bias in DL is often at a ""sweet spot"" which implicitly allows ANNs to learn well, due to unknown key design choices. But performance typically degrades when these choices change. Therefore, understanding and controlling algorithmic bias is the key to unlocking the true potential of deep learning.Our goal is to develop a rigorous theory of algorithmic bias in DL and to apply it to alleviate critical practical bottlenecks that prevent such models from scaling up or implemented in real-world applications.
Our approach has three objectives: (1) identify the algorithmic biases affecting DL; (2) understand how these biases affect the functional capabilities and generalization performance; and (3) control these biases to alleviate critical practical bottlenecks. To demonstrate the feasibility of this challenging project, we describe how recent advances and concrete preliminary results enable us to effectively approach all these objectives."
Status
SIGNEDCall topic
ERC-2021-STGUpdate Date
09-02-2023
Images
No images available.
Geographical location(s)
