Max Welling
29Patents
1h-index
60Co-inventors
60Inventor score
Filing activity: Feb 1, 2002 → Sep 30, 2022
Most-cited inventions
| Patent | Title | Area | Cited by | Status |
|---|---|---|---|---|
| US7280697B2 | Unsupervised learning of object categories from cluttered images | Physics | 17 | Expired |
| US11551759B2 | Voltage offset for compute-in-memory architecture | Physics | 1 | Active |
| US10885467B2 | Differentially private iteratively reweighted least squares | Physics | 1 | Active |
| US11616666B2 | Neural network augmentation for wireless channel estimation and tracking | Electricity | 1 | Active |
| US12200660B2 | Unsupervised learning for simultaneous localization and mapping in deep neural networks using channel state information | Physics | 1 | Active |
| US11790241B2 | Systems and methods for modifying neural networks for binary processing applications | Physics | 1 | Active |
| US12024159B2 | Device and method for generating a compressed network from a trained neural network | Physics | 0 | Active |
| US11696093B2 | Multi-object positioning using mixture density networks | Physics | 0 | Active |
| US11481649B2 | Adapting a base classifier to novel classes | Physics | 0 | Active |
| US11804034B2 | Training a function to respond predictably to differences | Physics | 0 | Active |
| US11836572B2 | Quantum inspired convolutional kernels for convolutional neural networks | Physics | 0 | Active |
| US12158922B2 | Gauge equivariant geometric graph convolutional neural network | Physics | 0 | Active |
| US11823302B2 | Device for and computer implemented method of digital signal processing | Physics | 0 | Active |
| US11562212B2 | Performing XNOR equivalent operations by adjusting column thresholds of a compute-in-memory array | Physics | 0 | Active |
| US11995151B2 | Image generation model based on log-likelihood | Physics | 0 | Active |
| US11620517B2 | Control of a physical system based on inferred state | Physics | 0 | Active |
| US12164302B2 | Configuring a neural network for equivariant or invariant behavior | Physics | 0 | Active |
| US12354230B2 | System and method with spherical autoencoder for object discovery | Physics | 0 | Active |
| US12248878B2 | Device and method for training a neuronal network | Physics | 0 | Active |
| US11842279B2 | Combinatorial Bayesian optimization using a graph cartesian product | Physics | 0 | Active |
| US11961275B2 | Device and method for training a normalizing flow | Physics | 0 | Active |
| US11700070B2 | Hypernetwork Kalman filter for channel estimation and tracking | Electricity | 0 | Active |
| US12269177B2 | Machine learning based on a probability distribution of sensor data | Physics | 0 | Active |
| US12100198B2 | Physical environment interaction with an equivariant policy | Physics | 0 | Active |
| US12412085B2 | Data and compute efficient equivariant convolutional networks | Physics | 0 | Active |
Source: USPTO / EPO open patent data. Inventor disambiguation is heuristic; counts are objective bibliographic measures.