Zahra Atashgahi

Ph.D. candidate @Utwente

[Zahra Atashgahi @ZAtashgahi]

Very happy to share that our paper "Supervised Feature Selection with Neuron Evolution in Sparse Neural Networks" has been accepted to TMLR. Thanks to my co-authors Xuhao, Neil, @ShiweiLiu9, @luu_yinn, @pechenizkiy, Raymond, and Decebal.

[Accepted papers at TMLR @TmlrPub]

Supervised Feature Selection with Neuron Evolution in Sparse Neural Networks

Zahra Atashgahi, Xuhao Zhang, Neil Kichler et al..

Action editor: Makoto Yamada.

https://t.co/E6DDjpjZh9

#features #neurofs #feature

[Data and AI cluster @ TU/e @Data_AI_TUe]

With 11 contributions, our cluster members will be very busy in Vilnius 🥳 See you soon at #ECMLPKDD 2024!

[Samaneh Khoshrou @Samakhosh]

We are looking for an ambitious and enthusiastic researcher (4-year PhD) who has a passion for artificial intelligence and law. For more information, check https://t.co/0Otne4JFzG and apply before the end of 2023 if you are interested!#human_centeredAI #NLP #CIVIC_AI

[Zahra Atashgahi @ZAtashgahi]

@g_sokar Congrats, Ghada🥳

[Zahra Atashgahi @ZAtashgahi]

@negin__ghasemi @SIGIRConf @maliannejadi @hamedrab @ekanou @arjenpdevries @jallan_umass @djoerd Congratulations Negin!😊

[Zahra Atashgahi @ZAtashgahi]

@NicStrisc @ShiweiLiu9 @luu_yinn @pechenizkiy Thanks, Nicola!😊

[Zahra Atashgahi @ZAtashgahi]

Very happy to share that our paper "Supervised Feature Selection with Neuron Evolution in Sparse Neural Networks" has been accepted to TMLR. Thanks to my co-authors Xuhao, Neil, @ShiweiLiu9, @luu_yinn, @pechenizkiy, Raymond, and Decebal.

[Accepted papers at TMLR @TmlrPub]

Supervised Feature Selection with Neuron Evolution in Sparse Neural Networks

Zahra Atashgahi, Xuhao Zhang, Neil Kichler et al..

Action editor: Makoto Yamada.

https://t.co/E6DDjpjZh9

#features #neurofs #feature

[Zahra Atashgahi @ZAtashgahi]

@Elena__Mocanu @ShiweiLiu9 @luu_yinn @pechenizkiy Thanks, Elena!

[Zahra Atashgahi @ZAtashgahi]

@ShiweiLiu9 Congratulations Shiwei🎉

[sparsenn @sparsenn]

📢The call for papers of the “Sparsity in Neural Networks: On practical limitations and tradeoffs between sustainability and efficiency” Workshop is online at https://t.co/lYkj8mkRYY.
🗓️Deadline: February 3, 2023, AoE.
#ICLR2023

[sparsenn @sparsenn]

Thank you ICLR for having us! Excited to meet with you all in Kigali this May. Stay tuned for more details!

[Zahra Atashgahi @ZAtashgahi]

@g_sokar Congrats and Happy birthday!😊🥳

[Zahra Atashgahi @ZAtashgahi]

@BramGrooten @aamas2023 @g_sokar @s_dohare @Elena__Mocanu @pechenizkiy Congrats Bram!🎉

[Zahra Atashgahi @ZAtashgahi]

We carried out different experiments on eight datasets, including tabular, image, and text datasets, and demonstrated that our proposed method outperforms several state-of-the-art sparse training algorithms in extremely sparse MLPs by a large gap.

[Zahra Atashgahi @ZAtashgahi]

A bit late announcement, our paper "A brain-inspired algorithm for training highly sparse neural networks" got published in Machine Learning Journal, ECML-PKKDD journal track. Thanks to co-authors: Joost Pieterse, @ShiweiLiu9, Decebal Mocanu, Raymond Veldhuis, and @pechenizkiy.

[Zahra Atashgahi @ZAtashgahi]

Concretely, by exploiting the cosine similarity metric to measure the importance of the connections, our proposed method, “Cosine similarity-based and Random Topology Exploration (CTRE),” evolves the topology of sparse neural networks by adding the most important connections.

[Zahra Atashgahi @ZAtashgahi]

@ShiweiLiu9 @LogConference @Tianjinhuang1 @luu_yinn @pechenizkiy @PetarV_93 Congratulations!🎉