AMA with ICML Authors

ICML is one of the top conferences in machine learning. Their 35th conference will happen in Stockholm on the 10th -15th July. After a rigorous review process, over 600 papers have been accepted to the conference.

As part of our ICML reading challenge, we will be having AMA sessions with authors of the following ICML accepted papers:

1. Asynchronous Byzantine Machine Learning (Georgios Damaskinos, El Mahdi El Mhamdi, Rachid Guerraoui, Rhicheek Patra, Mahsa Taziki)
2. Learning the Reward Function for a Misspecified Model (Erik Talvitie)
3. Inspectre: Privately Estimating the Unseen (Jayadev Acharya, Gautam Kamath, Ziteng Sun, Huanyu Zhang)
4. Learning Longer-term Dependencies in RNNs with Auxiliary Losses (Trieu Trinh, Andrew Dai, Thang Luong, Quoc Le)
5. Towards Fast Computation of Certified Robustness for ReLU Networks (Tsui-Wei Weng, Huan Zhang, Hongge Chen, Zhao Song, Cho-Jui Hsieh, Luca Daniel, Duane Boning, Inderjit Dhillon)
6. Augment and Reduce: Stochastic Inference for Large Categorical Distributions (Francisco J. R. Ruiz, Michalis K. Titsias, Adji B. Dieng, David M. Blei)

How does the AMA work

The AMA is not live. You can post a question or comment on the papers anytime and anywhere before 15 June (the earlier you post, the more likely your question will be answered).

Here is how you post a question:

1. Click on the paper’s link above
2. Highlight any text on the paper to form a public discussion thread
3. Post your questions/comments in the comment box. It could be on:

• Difficulties in replicating experimental results
• Terms and concepts that are not sufficiently elaborated
• How ideas in the paper can be extended further

Highlighting and creating a public discussion thread

Do you have a question on an ICML paper not listed above? Let us know and we will attempt to contact the author for an AMA.

This is your opportunity to level up your AI knowledge, don’t miss it.