An official website of the United States government
Contrastive learning (CL) has been widely investigated with various learning mech- anisms and achieves strong capability in learning representations of data in a self-supervised manner using unlabeled data. A common fashion of contrastive learning on this line is employing large-sized encoders to achieve comparable performance as the supervised learning counterpart. Despite the success of the labelless training, current contrastive learning algorithms failed to achieve good performance with lightweight (compact) models, e.g., MobileNet, while the re- quirements of the heavy encoders impede the energy-efficient computation, espe- cially for resource-constrained AI applications. Motivated by this, we propose a new self-supervised CL scheme, named SACL-XD, consisting of two technical components, Slimmed Asymmetrical Contrastive Learning (SACL) and Cross- Distillation (XD), which collectively enable efficient CL with compact models. While relevant prior works employed a strong pre-trained model as the teacher of unsupervised knowledge distillation to a lightweight encoder, our proposed method trains CL models from scratch and outperforms them even without such an expensive requirement. Compared to the SoTA lightweight CL training (dis- tillation) algorithms, SACL-XD achieves 1.79% ImageNet-1K accuracy improve- ment on MobileNet-V3 with 64⇥ training FLOPs reduction. Code is available at https://github.com/mengjian0502/SACL-XD.
more »
« less
This content will become publicly available on January 22, 2026
Self-trainable and adaptive sensor intelligence for selective data generation
With the increasing integration of machine learning into IoT devices, managing energy consumption and data transmission has become a critical challenge. Many IoT applications depend on complex computations performed on server-side infrastructure, necessitating efficient methods to reduce unnecessary data transmission. One promising solution involves deploying compact machine learning models near sensors, enabling intelligent identification and transmission of only relevant data frames. However, existing near-sensor models lack adaptability, as they require extensive pre-training and are often rigidly configured prior to deployment. This paper proposes a novel framework that fuses online learning, active learning, and knowledge distillation to enable adaptive, resource-efficient near-sensor intelligence. Our approach allows near-sensor models to dynamically fine-tune their parameters post-deployment using online learning, eliminating the need for extensive pre-labeling and training. Through a sequential training and execution process, the framework achieves continuous adaptability without prior knowledge of the deployment environment. To enhance performance while preserving model efficiency, we integrate knowledge distillation, enabling the transfer of critical insights from a larger teacher model to a compact student model. Additionally, active learning reduces the required training data while maintaining competitive performance. We validated our framework on both benchmark data from the MS COCO dataset and in a simulated IoT environment. The results demonstrate significant improvements in energy efficiency and data transmission optimization, highlighting the practical applicability of our method in real-world IoT scenarios.
more »
« less
- PAR ID:
- 10586591
- Publisher / Repository:
- Frontiers
- Date Published:
- Journal Name:
- Frontiers in Artificial Intelligence
- Volume:
- 7
- ISSN:
- 2624-8212
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
Continual learning is a setting where machine learning models learn novel concepts from continuously shifting training data, while simultaneously avoiding degradation of knowledge on previously seen classes which may disappear from the training data for extended periods of time (a phenomenon known as the catastrophic forgetting problem). Current approaches for continual learning of a single expanding task (aka class-incremental continual learning) require extensive rehearsal of previously seen data to avoid this degradation of knowledge. Unfortunately, rehearsal comes at a cost to memory, and it may also violate data-privacy. Instead, we explore combining knowledge distillation and parameter regularization in new ways to achieve strong continual learning performance without rehearsal. Specifically, we take a deep dive into common continual learning techniques: prediction distillation, feature distillation, L2 parameter regularization, and EWC parameter regularization. We first disprove the common assumption that parameter regularization techniques fail for rehearsal-free continual learning of a single, expanding task. Next, we explore how to leverage knowledge from a pre-trained model in rehearsal-free continual learning and find that vanilla L2 parameter regularization outperforms EWC parameter regularization and feature distillation. Finally, we explore the recently popular ImageNet-R benchmark, and show that L2 parameter regularization implemented in self-attention blocks of a ViT transformer outperforms recent popular prompting for continual learning methods.more » « less
-
A central challenge in machine learning deployment is maintaining accurate and updated models as the deployment environment changes over time. We present a hardware/software framework for simultaneous training and inference for monocular depth estimation on edge devices. Our proposed framework can be used as a hardware/software co-design tool that enables continual and online federated learning on edge devices. Our results show real-time training and inference performance, demonstrating the feasibility of online learning on edge devices.more » « less
-
Scalable methods for optical transmission performance prediction using machine learning (ML) are studied in metro reconfigurable optical add-drop multiplexer (ROADM) networks. A cascaded learning framework is introduced to encompass the use of cascaded component models for end-to-end (E2E) optical path prediction augmented with different combinations of E2E performance data and models. Additional E2E optical path data and models are used to reduce the prediction error accumulation in the cascade. Off-line training (pre-trained prior to deployment) and transfer learning are used for component-level erbium-doped fiber amplifier (EDFA) gain models to ensure scalability. Considering channel power prediction, we show that the data collection process of the pre-trained EDFA model can be reduced to only 5% of the original training set using transfer learning. We evaluate the proposed method under three different topologies with field deployed fibers and achieve a mean absolute error of 0.16 dB with a single (one-shot) E2E measurement on the deployed 6-span system with 12 EDFAs.more » « less
-
Understanding thermal stress evolution in metal additive manufacturing (AM) is crucial for producing high-quality components. Recent advancements in machine learning (ML) have shown great potential for modeling complex multiphysics problems in metal AM. While physics-based simulations face the challenge of high computational costs, conventional data-driven ML models require large, labeled training datasets to achieve accurate predictions. Unfortunately, generating large datasets for ML model training through time-consuming experiments or high-fidelity simulations is highly expensive in metal AM. To address these challenges, this study introduces a physics-informed neural network (PINN) framework that incorporates governing physical laws into deep neural networks (NNs) to predict temperature and thermal stress evolution during the laser metal deposition (LMD) process. The study also discusses enhanced accuracy and efficiency of the PINN model when supplemented with small simulation data. Furthermore, it highlights the PINN transferability, enabling fast predictions with a set of new process parameters using a pre-trained PINN model as an online soft sensor, significantly reducing computation time compared to physics-based numerical models while maintaining accuracy.more » « less
