Ensemble based discriminative models for Visual Dialog Challenge 2018
arXiv:2001.05865v11 citations
Originality Synthesis-oriented
AI Analysis
This work addresses visual dialog tasks for AI researchers, but it is incremental as it applies existing ensemble methods to a specific challenge.
The authors tackled the Visual Dialog Challenge 2018 by using an ensemble of three discriminative models, achieving an NDCG score of 55.46 and an MRR of 63.77, which secured third place in the competition.
This manuscript describes our approach for the Visual Dialog Challenge 2018. We use an ensemble of three discriminative models with different encoders and decoders for our final submission. Our best performing model on 'test-std' split achieves the NDCG score of 55.46 and the MRR value of 63.77, securing third position in the challenge.