TPGR: Large-scale Interactive Recommendation with Tree-structured Policy Gradient
Nov. 2018
Reinforcement Learning has recently been introduced to recommender systems. As Recommender systems are always with thousands of items to recommend, most existing RL-based methods fail to handle such a large discrete action space. We propose a Tree-structured Policy Gradient Recommendation (TPGR) framework, where a balanced hierarchical clustering tree is built over the items and picking an item is formulated as seeking a path from the root to a certain leaf.
View Detail
Noah helps the HUAWEI Mate 20 series lead the new height of wisdom
Oct. 2018
The HUAWEI Mate 20 series mobile phone, which was released in Shanghai on October 26th, has brought new wisdom to the world with many new technologies, attracting the attention of users all over the world. Noah's AI algorithm has become one of the engines that help the Mate 20 series achieve the new height of wisdom.
View Detail
Huawei Noah's Ark Lab has 6 papers accepted by NIPS 2018, including an Oral
Oct. 2018
NIPS, one of the top conferences in the field of machine learning, recently announced the list of papers accepted by NIPS 2018. Six papers from Huawei Noah's Ark Lab were accepted by NIPS, one of which was selected as an oral report (Oral, 0.6%, 30/4856). Among them, the first author of four papers is from Noah. In the industrial ranking based on the number of first author papers, Noah is the 7th in the world and the first in China.
View Detail
PIN: A More Advanced Deep Learning Model for Recommender System From Noah’s Ark Lab
Aug. 2018
Recently, deep neural networks have attracted research attention on such a problem for their high capacity and end-to-end training scheme. We propose kernel product to learn field-aware feature interactions and product-based Neural Network (PNN) which adopts a feature extractor to explore feature interactions. Generalizing the kernel product to a net-in-net architecture, we further propose Product-network In Network (PIN) which can generalize previous models.
View Detail
Professor Qun Liu Joined Huawei Noah's Ark Lab as the Chief Scientist of Speech and Language Computing
Jul. 2018
Dr. Qun Liu, a professor in Dublin City University, recently joined Huawei Noah's Ark Lab as the Chief Scientist of Speech and Language Computing, to lead cutting-edge research and technological innovation in the field of speech and natural language processing.
View Detail
Huawei Noah’s Ark Lab attended the CVPR conference
Jun. 2018
The top conference in the area of computer vision, CVPR (Computer vision and pattern recognition) was held in Salt Lake city, Utah State in the US, from June 18 to Jun22, 2018. Experts from the computer vision group in Huawei Noah’s Ark Lab attended the conference. Together with leads from both the industry and the academic, they discussed the development of computer vision and explore the future of the field.
View Detail
Computer vision expert Qi Tian joins Huawei Noah's Ark Laboratory as a chief scientist of computer vision
Jun. 2018
Recently, the computer vision expert Dr. Qi Tian, joined Huawei Noah’s Ark Laboratory as the chief scientist of computer vision. Dr. Tian was a Full Professor in the Department of Computer Science, the University of Texas at San Antonio (UTSA) before he took the position in Huawei Noah’s Ark Lab. He is an IEEE fellow, ...
View Detail
A Data-Driven Three-Layer Algorithm for Split Delivery Vehicle Routing Problem with 3D Container Loading Constraint
Jun. 2018
Split Delivery Vehicle Routing Problem with 3D Loading Constraints (3L-SDVRP) comprises three NP-hard problems, vehicle routing problem, cargo splitting problem and 3D container loading problem, which shall be jointly optimized for cost savings. We design a novel data-driven three-layer search algorithm (DTSA), which improves both the efficiency and effectiveness of traditional ...
View Detail
The Relationship Between Machine Learning and Operations Research
May. 2018
As an intern at Huawei's Noah's Ark Lab, my work is combining machine learning with operations research to improve the efficiency of Huawei's supply chain. It is found that the traditional OR methods, once combined with “learning”, will produce valuable research. And it's also one of the hottest research areas in solving large-scale complex problems. In this article, I will talk about my understandings about them.
View Detail
Attend VALSE 2018 to Explore the Future of Computational Vision
Apr. 2018
On April 22, 2018, the three-day 8th Vision And Learning SEminar (VALSE 2018) closed successfully in Dalian, a beautiful seaside city. Noah's Ark Lab participated in this grand event with numerous research results in the field of computational vision and had extensive and in-depth exchanges with the participants.
View Detail
DeepFM: A Deep Learning Model for App Recommendation in Huawei App Store
Apr. 2018
Learning sophisticated feature interactions behind user behaviors is critical for recommender systems. Despite great progress, existing methods have a strong bias towards low- and high-order interactions, or require expertise feature engineering. We derive an end-to-end learning model that emphasizes both low- and high-order feature interactions. The proposed model, DeepFM, combines the power of factorization machines for recommendation and deep learning for feature learning in a new neural network architecture.
View Detail
Deep Meta-Learning: Learning to Learn in the Concept Space
Apr. 2018
We propose a new meta-learning framework to integrate the representation power of deep learning into meta-learning, thus called deep meta-learning, and show it can substantially improve vanilla meta-learning algorithms on various few-shot image recognition problems. For example, on 5-way-1-shot image recognition on CUB-200, it improves Meta-SGD (Li et al., 2017) from 53.34% to 66.95%. We expect its widespread use in the near future.
View Detail
Prof. Thomas Dietterich made a visit to Noah’s Ark Lab
May. 2017
Prof. Dietterich is a leader of research on Robust AI, which is about how to make AI systems more robust in practice, particularly in high-risk applications. Researchers at Noah’s Ark are collaborating with Prof. Dietterich on Robust AI with machine learning for telecommunications as the application area....
View Detail