Industry
CLOOB: Modern Hopfield Networks with InfoLOOB Outperform CLIP
Andreas Fürst, Elisabeth Rumetshofer, Johannes Lehner, Viet Tran, Fei Tang, Hubert Ramsauer, David Kreil, Michael Kopp, Günter Klambauer, Angela Bitto-Nemling, Sepp Hochreiter
CLOOB: Modern Hopfield Networks with InfoLOOB Outperform CLIP
CLOOB: Modern Hopfield Networks with InfoLOOB Outperform CLIP
Andreas Fürst, Elisabeth Rumetshofer, Johannes Lehner, Viet Tran, Fei Tang, Hubert Ramsauer, David Kreil, Michael Kopp, Günter Klambauer, Angela Bitto-Nemling, Sepp Hochreiter
CLOOB: Modern Hopfield Networks with InfoLOOB Outperform CLIP
High-ThroughputSynchronousDeepRL
Deep reinforcement learning (RL) is computationally demanding and requiresprocessing of many data points. Synchronous methods enjoy training stability while having lowerdatathroughput.
Review for NeurIPS paper: Combining Deep Reinforcement Learning and Search for Imperfect-Information Games
The paper was reviewed by experts on the topic and discussed after authors rebuttal. Results were found to be interesting and valuable. The reviewers comments should be taken into account while preparing the final version of the paper.
