Alireza Furutanpey
PulseAugur coverage of Alireza Furutanpey — every cluster mentioning Alireza Furutanpey across labs, papers, and developer communities, ranked by signal.
2 day(s) with sentiment data
-
New SG-AFM method enhances learned image compression by 13%
Researchers have developed a new method called Semantic-Guided Adaptive Latent-Frequency Modulation (SG-AFM) to improve learned image compression. This technique uses semantic information to guide the attenuation of hig…
-
New paper explores randomized smoothing geometry on feasible sets
Researchers Alireza Furutanpey and Ruevo ny have published a paper on arXiv detailing the geometry of randomized smoothing on feasible sets. Their work introduces a method to certify label probabilities by filtering pro…
-
FANS framework optimizes model architectures for heterogeneous federated learning
Researchers have developed FANS (Federated Adaptive Network Search), a new framework designed to optimize model architectures in heterogeneous federated learning environments. This approach utilizes a hypernetwork to le…
-
New neural codec optimizes satellite image transmission by prioritizing clear content
Researchers have developed a novel neural codec designed to improve the efficiency of transmitting imagery from Earth-observation satellites. This codec utilizes a clear-probability-weighted reconstruction loss, allowin…
-
New intrinsic motivation method for reinforcement learning unveiled
Researchers have introduced a novel approach to intrinsic motivation in unsupervised reinforcement learning, focusing on environments with mixed uncertainty sources. The proposed method, termed "Principled Direction-Fre…
-
Research paper evaluates neural graph compilers' impact on ML performance
This research paper investigates the impact of neural network graph compilers on the performance of machine learning models across various hardware platforms. The study highlights how vendor-specific optimizations can s…
-
Research questions RL's real-world gains in service orchestration
A new research paper questions the widespread adoption of reinforcement learning (RL) for service orchestration, arguing that publication incentives favor benchmark gains over real-world performance evidence. The study …