New Awesome Papers, Books, and Libraries?

[o-laurent]

Here is the place for you to propose papers or everything else related to uncertainty in deep learning!

[wangyucheng1234]

Thank you for maintaining this excellent repository! It has been a valuable resource for researchers working on uncertainty quantification problems in ML like me.

I would like to kindly suggest adding our recent paper to the collection:

1. Hashing with Uncertainty Quantification via Sampling-based Hypothesis Testing (TMLR):
   This paper develops a probabilistic hashing model(ProbHash). Sampling-based hypothesis testing is then derived for hashing with uncertainty quantification(HashUQ) in ProbHash to improve the granularity of hashing-based retrieval by prioritizing the data with confident hash-codes.

2. Uncertainty-aware Unsupervised Video Hashing (AISTATS23):
   This paper proposes to formulate video hashing as learning a discrete variational auto-encoder with the factorized Bernoulli latent distribution, which enables uncertainty-aware video hashing by predicting the probability distribution of video hash code-words, thus providing reliable uncertainty quantification.

Thank you for considering my suggestion, and please let me know if you need any additional information!

Best regards,
Yucheng Wang

[xuanlongORZ]

Thanks for your appreciation and contribution; I suggest placing the two papers in Bayesian-methods section if you agree. You can also send a pull request and do it by yourself.

Best,
Xuanlong

[wangyucheng1234]

Thank you for your reply!

I agree that these two papers could potentially fit well under the Bayesian-methods section. Please let me know if there’s any input you’d like from me.

Best regards,
Yucheng