Zobrazeno 1 - 10
of 35
pro vyhledávání: '"Jha, Nandan Kumar"'
Autor:
Jha, Nandan Kumar, Reagen, Brandon
The pervasiveness of proprietary language models has raised privacy concerns for users' sensitive data, emphasizing the need for private inference (PI), where inference is performed directly on encrypted inputs. However, current PI methods face prohi
Externí odkaz:
http://arxiv.org/abs/2410.13060
Autor:
Jha, Nandan Kumar, Reagen, Brandon
LayerNorm is a critical component in modern large language models (LLMs) for stabilizing training and ensuring smooth optimization. However, it introduces significant challenges in mechanistic interpretability, outlier feature suppression, faithful s
Externí odkaz:
http://arxiv.org/abs/2410.09637
Autor:
Jha, Nandan Kumar, Reagen, Brandon
Prior work on Private Inference (PI) -- inferences performed directly on encrypted input -- has focused on minimizing a network's ReLUs, which have been assumed to dominate PI latency rather than FLOPs. Recent work has shown that FLOPs for PI can no
Externí odkaz:
http://arxiv.org/abs/2304.10593
In two-party machine learning prediction services, the client's goal is to query a remote server's trained machine learning model to perform neural network inference in some application domain. However, sensitive information can be obtained during th
Externí odkaz:
http://arxiv.org/abs/2207.07177
The privacy concerns of providing deep learning inference as a service have underscored the need for private inference (PI) protocols that protect users' data and the service provider's model using cryptographic methods. Recently proposed PI protocol
Externí odkaz:
http://arxiv.org/abs/2111.02583
Privacy concerns in client-server machine learning have given rise to private inference (PI), where neural inference occurs directly on encrypted inputs. PI protects clients' personal data and the server's intellectual property. A common practice in
Externí odkaz:
http://arxiv.org/abs/2107.12342
The simultaneous rise of machine learning as a service and concerns over user privacy have increasingly motivated the need for private inference (PI). While recent work demonstrates PI is possible using cryptographic primitives, the computational ove
Externí odkaz:
http://arxiv.org/abs/2106.08475
The recent rise of privacy concerns has led researchers to devise methods for private neural inference -- where inferences are made directly on encrypted data, never seeing inputs. The primary challenge facing private inference is that computing on e
Externí odkaz:
http://arxiv.org/abs/2103.01396
Autor:
Jha, Nandan Kumar, Mittal, Sparsh
In recent years, researchers have focused on reducing the model size and number of computations (measured as "multiply-accumulate" or MAC operations) of DNNs. The energy consumption of a DNN depends on both the number of MAC operations and the energy
Externí odkaz:
http://arxiv.org/abs/2008.02565
The remarkable predictive performance of deep neural networks (DNNs) has led to their adoption in service domains of unprecedented scale and scope. However, the widespread adoption and growing commercialization of DNNs have underscored the importance
Externí odkaz:
http://arxiv.org/abs/2007.15248