This inventor holds 1 USPTO granted patent. Top assignees: International Business Machines Corporation, Purdue Research Foundation. Active years: 2026.
Data source: USPTO (United States Patent and Trademark Office) public patent records. Weekly synchronization. How IDiyas builds this profile
Company Filing History:


Years Active: 2026
Title: Ankush Singla: Inventor in Machine Learning
Introduction
Ankush Singla is an inventor associated with Hayward, California. His patent work addresses machine learning methods for transferring knowledge between domains.
Latest Patents
Singla is credited with 1 patent: “Autoencoder with Generative Adversarial Networks for Transfer Learning Between Domains.” The invention combines a generator, encoder, decoder, discriminator, and classifier. The generator creates a domain-independent representation that can support classification and reconstruction while attempting to fool the discriminator. The encoder also produces a domain-dependent representation, which is constrained to contain little information. Together, the representations retain enough information to reconstruct the input sample while supporting transfer learning across domains.
Career Highlights
The supplied career information associates Singla with International Business Machines Corporation and Purdue Research Foundation.
Collaborations
Yuhai Tu is identified as a coworker.
Conclusion
Ankush Singla’s patent describes an approach to domain transfer that balances domain-independent learning with input reconstruction, classification, and limited domain-specific information.
Data source: USPTO (United States Patent and Trademark Office) public patent records. Weekly synchronization. How IDiyas builds this profile