Researchers at National Taiwan University are developing low-complexity neural network technologies using quantization to reduce model size while maintaining accuracy. Their work includes binary-weighted CNNs and transformers, along with a neural architecture search scheme (TPC-NAS) applied to image recognition, object detection, and NLP tasks. They have also built a PE-based CNN/transformer hardware accelerator in Xilinx FPGA SoC with a PyTorch-based software framework. Why it matters: This research provides practical methods for deploying efficient deep learning models on resource-constrained hardware, potentially enabling broader adoption of AI in embedded systems and edge devices.
This paper analyzes the impact of device uncertainties on deep neural networks (DNNs) in emerging device-based Computing-in-memory (CiM) systems. The authors propose UAE, an uncertainty-aware Neural Architecture Search scheme, to identify DNN models robust to these uncertainties. The goal is to mitigate accuracy drops when deploying trained models on real-world platforms.
MBZUAI researchers are introducing MedNNS, a system to be presented at MICCAI 2025, that recommends the best AI architecture and initialization for medical imaging tasks. MedNNS addresses the challenge of inefficient trial-and-error in building medical imaging AI by reframing model selection as a retrieval problem. The system employs a Once-For-All ResNet-like model and a learned meta-space of 720k model-dataset pairs, using dataset embeddings to predict optimal model performance. Why it matters: By automating model selection, MedNNS promises to significantly reduce the time and resources required to develop effective AI solutions for healthcare, particularly in medical imaging.
The paper introduces MedNNS, a neural network search framework designed for medical imaging, addressing challenges in architecture selection and weight initialization. MedNNS constructs a meta-space encoding datasets and models based on their performance using a Supernetwork-based approach, expanding the model zoo size by 51x. The framework incorporates rank loss and Fréchet Inception Distance (FID) loss to capture inter-model and inter-dataset relationships, improving alignment in the meta-space and outperforming ImageNet pre-trained DL models and SOTA NAS methods.
A new neural network architecture called Orchid was introduced that uses adaptive convolutions to achieve quasilinear computational complexity O(N logN) for sequence modeling. Orchid adapts its convolution kernel dynamically based on the input sequence. Evaluations across language modeling and image classification show that Orchid outperforms attention-based architectures like BERT and Vision Transformers, often with smaller model sizes. Why it matters: Orchid extends the feasible sequence length beyond the practical limits of dense attention layers, representing progress toward more efficient and scalable deep learning models.
Nvidia is expanding its market beyond GPUs with the development of a central processing unit (CPU) based on Arm architecture. This move positions Nvidia to compete directly with established CPU manufacturers like Intel and AMD. The company aims to offer integrated hardware and software solutions optimized for AI and data science workloads. Why it matters: Nvidia's entry into the CPU market could accelerate AI development and adoption in the Gulf region by providing more specialized and efficient computing solutions.
The article by RSM US LLP discusses the process of establishing an AI strategy and an implementation plan tailored to an organization's specific needs. It likely covers key considerations for integrating AI, such as identifying business objectives, assessing current capabilities, and developing a roadmap for adoption. The publication aims to guide organizations in developing a coherent approach to leverage artificial intelligence effectively within their operations. Why it matters: While general in scope, frameworks for AI strategy implementation are foundational for organizations in the Middle East as they develop their own AI roadmaps.