相关论文: Extending the Lora modulation to add further paral…

Little research explores the correlation between the expressive ability and generalization ability of the low-rank adaptation (LoRA). Sharpness-Aware Minimization (SAM) improves model generalization for both Convolutional Neural Networks…

Low-rank adaptation (LoRA) and its mixture-of-experts (MOE) variants are highly effective parameter-efficient fine-tuning (PEFT) methods. However, they introduce significant latency in multi-tenant settings due to the LoRA modules and MOE…

The Open Radio Access Network (RAN) paradigm is transforming cellular networks into a system of disaggregated, virtualized, and software-based components. These self-optimize the network through programmable, closed-loop control, leveraging…

We propose TLoRA, a novel tri-matrix low-rank adaptation method that decomposes weight updates into three matrices: two fixed random matrices and one trainable matrix, combined with a learnable, layer-wise scaling factor. This tri-matrix…

Integrating artificial intelligence (AI) into wireless networks has drawn significant interest in both industry and academia. A common solution is to replace partial or even all modules in the conventional systems, which is often lack of…

Increasing data volumes delivered by a new generation of radio interferometers require computationally efficient and robust calibration algorithms. In this paper, we propose distributed calibration as a way of improving both computational…

Low-Rank Adaptation (LoRA) and its variants have shown impressive results in reducing the number of trainable parameters and memory requirements of large transformer networks while maintaining fine-tuning performance. The low-rank nature of…

Propeller modulation, also known as micro-Doppler modulation, presents a significant challenge in radio frequency (RF) inspection operations conducted via drones. This paper investigates the equalization of propeller modulation effects on…

Parameter-Efficient Fine-Tuning (PEFT) methods like LoRA have significantly improved the adaptation of LLMs to downstream tasks in a resource-efficient manner. However, in multi-task scenarios, challenges such as training imbalance and the…

LoRa is a popular wireless technology that enables low-throughput (bytes) long-range communication (km) at low energy consumption (mW). Its transmission, though, is on one side prone to interference during long on-air times, and on the…

A resource-constrained unmanned aerial vehicle (UAV) can be used as a flying LoRa gateway (GW) to move inside the target area for efficient data collection and LoRa resource management. In this work, we propose deep reinforcement learning…

This paper investigates the use of deep reinforcement learning (DRL) in a MAC protocol for heterogeneous wireless networking referred to as Deep-reinforcement Learning Multiple Access (DLMA). The thrust of this work is partially inspired by…

It is well known that biology-inspired self-maintaining algorithms in wireless sensor nodes achieve near optimum time division multiple access (TDMA) characteristics in a decentralized manner and with very low complexity. We extend such…

Instruction Tuning has the potential to stimulate or enhance specific capabilities of large language models (LLMs). However, achieving the right balance of data is crucial to prevent catastrophic forgetting and interference between tasks.…

Radio frequency fingerprints (RFFs) enable secure wireless authentication but struggle in open-set scenarios with unknown devices and varying channels. Existing methods face challenges in generalization and incur high computational costs.…

This paper applies machine learning to optimize the transmission policy of cognitive radio inspired non-orthogonal multiple access (CR-NOMA) networks, where time-division multiple access (TDMA) is used to serve multiple primary users and an…

We theoretically and numerically investigate the evolution of discrete soliton in a 1D linearly chirped nonlinear waveguide array (WA). The discrete soliton is self-accelerated inside the transversely chirped WA and emits a \textit{dynamic…

This paper studies a specific low-power wireless technology capable of reaching a long range, namely LoRa. Such a technology can be used by different applications in cities involving many transmitting devices while requiring loose…

Low-rank adaptation (LoRA) is a natural method for finetuning in communication-constrained machine learning settings such as cross-device federated learning. Prior work that has studied LoRA in the context of federated learning has focused…

In this paper, we show that Low Rank Adaptation (LoRA) as originally introduced in Hu et al. (2021) leads to suboptimal finetuning of models with large width (embedding dimension). This is due to the fact that adapter matrices A and B in…