The results are based on anovelanalysis ofreal-time dynamic programming, thenextended tomodel-based RL.Specifically,wegeneralize existing algorithms that perform full-planning to act by 1-step planning.

Based oninvariant features, a high-performing classifier on source domains could hopefully behave equally well on a target domain. In other words, we hope the invariant features to be transferable. However, in practice, there are no perfectly transferable features, andsomealgorithmsseemtolearn"moretransferable"featuresthanothers.

Automating garment manipulation poses a significant challenge for assistive robotics due to the diverse and deformable nature of garments. Traditional approaches typically require separate models for each garment type, which limits scalability and adaptability. In contrast, this paper presents a unified approach using vision-language models (VLMs) to improve keypoint prediction across various garment categories. By interpreting both visual and semantic information, our model enables robots to manage different garment states with a single model. We created a large-scale synthetic dataset using advanced simulation techniques, allowing scalable training without extensive real-world data. Experimental results indicate that the VLM-based method significantly enhances keypoint detection accuracy and task success rates, providing a more flexible and general solution for robotic garment manipulation. In addition, this research also underscores the potential of VLMs to unify various garment manipulation tasks within a single framework, paving the way for broader applications in home automation and assistive robotics for future.

Test-time adaptation (TTA) is a task that continually adapts a pre-trained source model to the target domain during inference. One popular approach involves fine-tuning model with cross-entropy loss according to estimated pseudo-labels. However, its performance is significantly affected by noisy pseudo-labels. This study reveals that minimizing the classification error of each sample causes the cross-entropy loss's vulnerability to label noise. To address this issue, we propose a novel Decoupled Prototype Learning (DPL) method that features prototype-centric loss computation. First, we decouple the optimization of class prototypes. For each class prototype, we reduce its distance with positive samples and enlarge its distance with negative samples in a contrastive manner. This strategy prevents the model from overfitting to noisy pseudo-labels. Second, we propose a memory-based strategy to enhance DPL's robustness for the small batch sizes often encountered in TTA. We update each class's pseudo-feature from a memory in a momentum manner and insert an additional DPL loss. Finally, we introduce a consistency regularization-based approach to leverage samples with unconfident pseudo-labels. This approach transfers feature styles of samples with unconfident pseudo-labels to those with confident pseudo-labels. Thus, more reliable samples for TTA are created. The experimental results demonstrate that our methods achieve state-of-the-art performance on domain generalization benchmarks, and reliably improve the performance of self-training-based methods on image corruption benchmarks. The code will be released.

We study the problem of hanging a wide range of grasped objects on diverse supporting items. Hanging objects is a ubiquitous task that is encountered in numerous aspects of our everyday lives. However, both the objects and supporting items can exhibit substantial variations in their shapes and structures, bringing two challenging issues: (1) determining the task-relevant geometric structures across different objects and supporting items, and (2) identifying a robust action sequence to accommodate the shape variations of supporting items. To this end, we propose Semantic Keypoint Trajectory (SKT), an object-agnostic representation that is highly versatile and applicable to various everyday objects. We also propose Shape-conditioned Trajectory Deformation Network (SCTDN), a model that learns to generate SKT by deforming a template trajectory based on the task-relevant geometric structure features of the supporting items. We conduct extensive experiments and demonstrate substantial improvements in our framework over existing robot hanging methods in the success rate and inference time. Finally, our simulation-trained framework shows promising hanging results in the real world. For videos and supplementary materials, please visit our project webpage: https://hcis-lab.github.io/SKT-Hang/.

KDCA official Na Seong-woong (left) and SKT AI service head Lee Hyun-ah hold an MOU certificate at the KDCA headquarters in Cheongju, North Chungcheong Province. SK Telecom, South Korea's top telecom company, is utilizing its artificial intelligence technology to support the country's health authority in monitoring recipients after COVID-19 vaccination, it said Thursday. Dubbed NUGU Vaccine Carecall, SKT's NUGU AI platform will provide guidance to those subject to vaccination through calls, and monitor any abnormal signs after shots are administered. The telecom company signed a Memorandum of Understanding with the Korea Disease Control and Prevention Agency on Thursday. Under the MOU, medical institutions will register their lists of recipients on the NUGU Vaccine Carecall website.

SK Telecom (SKT) today unveiled its self-developed artificial intelligence (AI) chip named'SAPEON X220' and shared its AI semiconductor business vision. SAPEON X220 is optimally designed to process artificial intelligence tasks faster, using less power by efficiently processing large amounts of data in parallel. Its deep learning computation speed is 6.7 kilo-frames per second, which is 1.5 times faster than that of Graphics Processing Units (GPUs) for inference that are being widely used by AI-service companies. At the same time, it uses 20% less power than GPU by consuming 60 watts of energy and is about half the price of a GPU. SKT explained that SAPEON X220 will enable the provision of high-quality AI services by enhancing the performance of AI data centers through speedy computation of massive amounts of data.

SK Telecom is definitely a forward looking companies and will be one of the first companies to trial and start 5G. It is not wasting time when it comes to speech recognition and AI as SK Telecom has announced the first commercial adoption of FPGA accelerators from Xilinx in the AI domain for large scale data centers in South Korea. These two companies made some big claims over the efficiency of the FPGA based solution. SK Telecom has chosen the Xilinx Kintex UltraScale FPGAs as its artificial intelligence (AI) accelerators in its data center. Xilinx FPGA will take on the responsible task of running SKT's automatic speech-recognition (ASR) application to accelerate Nugu, SKT's voice-activated assistant.

South Korea's largest mobile operator SK Telecom (SKT) plans to invest KRW5 trillion ($4.1 billion) over the next three years to develop artificial intelligence (AI) and Internet of Things (IoT) technology and services. SKT said in a statement the investment strategy will focus on creating a "new ecosystem" for AI, self-driving cars and IoT, Yonhap news agency reported. Research will be coordinated with SKT's SK Broadband and SK Planet affiliates. A part of the investment will be allocated to support local start-ups, the operator said. At CES in Las Vegas last week, SKT said it held discussions with Samsung, Nvidia and Intel about cooperating on AI, connected cars and IoT initiatives, according to Yonhap.