Twitter Beğeni Hilesi

In this new Mila blog post, PhD candidate Ezgi Korkmaz proposes a framework dedicated to investigating the decision boundary and loss landscape similarities across states, MDPs and algorithms via adversarial directions. Read more 👉 pulse.ly/ct724tnx2x

If you are curious about deep reinforcement learning see my paper, published in AAAI, that reveals the intriguing differences between artificial and natural intelligence, and further juxtaposes robust training, adversarial attacks and generalization. ojs.aaai.org/index.php/AAAI… #AAAI

Excited to share that my paper on understanding and diagnosing deep reinforcement learning is accepted at #ICML2024. This paper introduces a foundational approach to analyze deep reinforcement learning decision making. ICML Conference

Very soon the paper will be online at #ICML2024. Truly excited to share these results! ICML Conference

Thank you so much for the kind words everyone. The paper is now online. #ICML2024 Link: openreview.net/pdf?id=s9RKqT7… x.com/EzgiKorkmazAI/…

Interested in reinforcement learning find below what UCL Blog Series published about one of my papers. blogs.ucl.ac.uk/steapp/2023/12… #NeurIPS2024 #ICML2024

Exciting times! I just saw that BSI cited one of my papers. Quite exciting to see the volume of recognition and interest. bsi.bund.de/SharedDocs/Dow… #ICML2024

The paper is now available on the #ICML2024 website! x.com/EzgiKorkmazAI/…

In case anyone would need this or would be interested, I am going to leave a recent survey I wrote about deep reinforcement learning here: Paper : arxiv.org/pdf/2401.02349 #NeurIPS2024 #ACL2024 #ICML2024

Which of these do you think will be more important in the next five years? #NeurIPS2024 #ICML2024 #ACL2024

This is quite interesting! Nice to see what the community thinks. x.com/EzgiKorkmazAI/… #ICML2024 #NeurIPS2024 #ACL2024

If you missed the poster session find the paper below: #ICML2024 ICML Conference x.com/EzgiKorkmazAI/…

Just realized that I have not shared this gem before! If you are interested in deep reinforcement learning find one of my recent #ICML papers below! ICML Conference Paper: proceedings.mlr.press/v202/korkmaz23……… #ICML2024 #AISafety #SafeRL #RobustRL UCL Computer Science