activity
20112022
most citedFlowNet: Learning Optical Flow with Convolutional Networks

604 citations · 1.4k across the 64 of their papers we have counts for

collaborators
Showing cs.LGShow all

21 papers · 1 filter

cs.LG2022

PRISM: Probabilistic Real-Time Inference in Spatial World Models

Atanas Mirchev, Baris Kayalibay, Ahmed Agha +3

We introduce PRISM, a method for real-time filtering in a probabilistic generative model of agent motion and visual perception. Previous approaches either lack uncertainty estimate…

cs.LG20221 cited

A Graph Is More Than Its Nodes: Towards Structured Uncertainty-Aware Learning on Graphs

Hans Hao-Hsun Hsu, Yuesong Shen, Daniel Cremers

Current graph neural networks (GNNs) that tackle node classification on graphs tend to only focus on nodewise scores and are solely evaluated by nodewise metrics. This limits uncer…

cs.LG20224 cited

Deep Combinatorial Aggregation

Yuesong Shen, Daniel Cremers

Neural networks are known to produce poor uncertainty estimations, and a variety of approaches have been proposed to remedy this issue. This includes deep ensemble, a simple and ef…

cs.LG202210 cited

What Makes Graph Neural Networks Miscalibrated?

Hans Hao-Hsun Hsu, Yuesong Shen, Christian Tomani +1

Given the importance of getting calibrated predictions and reliable uncertainty estimations, various post-hoc calibration methods have been developed for neural networks on standar…

cs.LG20221 cited

CHALLENGER: Training with Attribution Maps

Christian Tomani, Daniel Cremers

We show that utilizing attribution maps for training neural networks can improve regularization of models and thus increase performance. Regularization is key in deep learning, esp…

cs.LG2021

Dive into Layers: Neural Network Capacity Bounding using Algebraic Geometry

Ji Yang, Lu Sang, Daniel Cremers

The empirical results suggest that the learnability of a neural network is directly related to its size. To mathematically prove this, we borrow a tool in topological algebra: Bett…