Deep Learning
Graph Few-shot Learning with Task-specific Structures
Under the few-shot scenario, models are often required to conduct classification given limited labeled samples. Existing graph few-shot learning methods typically leverage Graph Neural Networks (GNNs) and perform classification across a series of meta-tasks. Nevertheless, these methods generally rely on the original graph (i.e., the graph that the meta-task is sampled from) to learn node
Overview
We provide additional details and results to complement the main paper. Fracture reassembly is an important task in the real world, e.g. We believe this research benefits both the economy and society. Even with advanced learning methods, human trust in AI is still a problem. Figure 6 shows the architecture of the three baseline methods, i.e., Global, LSTM and DGL. Then, we concatenate the global feature with each part feature and apply a shared-weight MLP network to regress the SE(3) pose for each input point cloud.
Breaking Bad: A Dataset for Geometric Fracture and Reassembly Silvia Sellán 1, Y un-Chun Chen 1,2, Ziyi Wu
We introduce Breaking Bad, a large-scale dataset of fractured objects.
Gossip-based Actor-Learner Architectures for Deep Reinforcement Learning
Mahmoud ("Mido") Assran, Joshua Romoff, Nicolas Ballas, Joelle Pineau, Mike Rabbat
Neural Information Processing Systems http://nips.cc/
Exploring evolution-aware &-free protein language models as protein function predictors
AlphaFold, Evoformer, has not been explored beyond structure prediction.
Learning Latent Seasonal-Trend Representations for Time Series Forecasting
However, clear patterns are still hard to extract since time series are often composed of several intricately entangled components.
Supplementary Material: A Transformer-Based Object Detector with Coarse-Fine Crossing Representations
The overall architecture of CFDT is shown in Figure 1. The base backbone is consistent with the network illustrated in the section of 3.1 Local-Global Cross Fusion. As shown by the red dotted lines in Figure 1, we use 100 det tokens as the additional input to perform self attention in the backbone. The det tokens dimension is also set as 256. Neck is a decoder-only modules, and there are 6 decoder layers in this neck.
