Deep Learning
Video Object Segmentation with Adaptive Feature Bank and Uncertain-Region Refinement
We introduced an adaptive feature bank update scheme to dynamically absorb new features and discard obsolete features.
Progressive Augmentation of GANs
Neural Information Processing Systems http://nips.cc/
Deep Temporal Sigmoid Belief Networks for Sequence Modeling
Zhe Gan, Chunyuan Li, Ricardo Henao, David E. Carlson, Lawrence Carin
Deep dynamic generative models are developed to learn sequential dependencies in time-series data. The multi-layered model is designed by constructing a hierarchy of temporal sigmoid belief networks (TSBNs), defined as a sequential stack of sigmoid belief networks (SBNs). Each SBN has a contextual hidden state, inherited from the previous SBNs in the sequence, and is used to regulate its hidden bias. Scalable learning and inference algorithms are derived by introducing a recognition model that yields fast sampling from the variational posterior. This recognition model is trained jointly with the generative model, by maximizing its variational lower bound on the log-likelihood. Experimental results on bouncing balls, polyphonic music, motion capture, and text streams show that the proposed approach achieves state-of-the-art predictive performance, and has the capacity to synthesize various sequences.
Self-Distillation Amplifies Regularization in Hilbert Space
Knowledge distillation introduced in the deep learning context is a method to transfer knowledge from one architecture to another. In particular, when the architectures are identical, this is called self-distillation. The idea is to feed in predictions of the trained model as new target values for retraining (and iterate this loop possibly a few times). It has been empirically observed that the self-distilled model often achieves higher accuracy on held out data.
An Adaptive Empirical Bayesian Method for Sparse Deep Learning
Wei Deng, Xiao Zhang, Faming Liang, Guang Lin
Neural Information Processing Systems http://nips.cc/
Ode to an ODE Krzysztof Choromanski Robotics at Google NY Jared Quincy Davis DeepMind & Stanford University V alerii Likhosherstov
This led to successful applications of Neural ODEs in machine learning.
Subspace Attack: Exploiting Promising Subspaces for Query-Efficient Black-box Attacks
Yiwen Guo, Ziang Yan, Changshui Zhang
Neural Information Processing Systems http://nips.cc/
'My son genuinely believed it was real': Parents are letting little kids play with AI. Are they wrong?
'My son genuinely believed it was real': Parents are letting little kids play with AI. Some believe AI can spark their child's imagination through personalized stories and generative images. Josh was at the end of his rope when he turned to ChatGPT for help with a parenting quandary. The 40-year-old father of two had been listening to his "super loquacious" four-year-old talk about Thomas the Tank Engine for 45 minutes, and he was feeling overwhelmed. "He was not done telling the story that he wanted to tell, and I needed to do my chores, so I let him have the phone," recalled Josh, who lives in north-west Ohio. "I thought he would finish the story and the phone would turn off."