To Label or Not to Label: Hybrid Active Learning for Neural Machine Translation

no code implementations • 14 Mar 2024 • Abdul Hameed Azeemi, Ihsan Ayyub Qazi, Agha Ali Raza

A weighted hybrid score that combines uncertainty and diversity is then used to select the top instances for annotation in each AL iteration.

Active Learning Domain Adaptation +4

Representative Subset Selection for Efficient Fine-Tuning in Self-Supervised Speech Recognition

no code implementations • 18 Mar 2022 • Abdul Hameed Azeemi, Ihsan Ayyub Qazi, Agha Ali Raza

Self-supervised speech recognition models require considerable labeled training data for learning high-fidelity representations for Automatic Speech Recognition (ASR) which is computationally demanding and time-consuming.

Active Learning Automatic Speech Recognition +2

FedPrune: Towards Inclusive Federated Learning

no code implementations • 27 Oct 2021 • Muhammad Tahir Munir, Muhammad Mustansar Saeed, Mahad Ali, Zafar Ayyub Qazi, Ihsan Ayyub Qazi

Federated learning (FL) is a distributed learning technique that trains a shared model over distributed data in a privacy-preserving manner.

Fairness Federated Learning +1

Countering Misinformation on Social Media Through Educational Interventions: Evidence from a Randomized Experiment in Pakistan

no code implementations • 6 Jul 2021 • Ayesha Ali, Ihsan Ayyub Qazi

Fake news is a growing problem in developing countries with potentially far-reaching consequences.

Misinformation