Xue (Steve) Liu

dheeraj.vattikonda@mila.quebec

Google Scholar

Sam Chen

Doctorat - McGill University

Co-superviseur⋅e :

Doctorat - McGill University

chengming.hu@mila.quebec

Dheeraj Vattikonda

Maîtrise recherche - McGill University

Dun Yuan

Maîtrise recherche - McGill University

dun.yuan@mila.quebec

Gaspard Zhao

Postdoctorat - McGill University

Co-superviseur⋅e :

Gregory Dudek

hanqing.zhao@mila.quebec

Mark Bai

Maîtrise recherche - McGill University

mark.bai@mila.quebec

Haolun Wu

Doctorat - McGill University

haolun.wu@mila.quebec

Junliang Luo

Doctorat - McGill University

junliang.luo@mila.quebec

rishabh.thaney@mila.quebec

Lyu Fuyuan

Doctorat - McGill University

Co-superviseur⋅e :

Jin Guo

fuyuan.lyu@mila.quebec

Maîtrise recherche - McGill University

shaoxiang.qin@mila.quebec

Shaoxiang Qin

Doctorat - McGill University

Shuhao Zheng

Doctorat - McGill University

shuhao.zheng@mila.quebec

Xuan Li

Doctorat - McGill University

xuan.li@mila.quebec

Ye Yuan

Doctorat - McGill University

Yili Jin

Doctorat - McGill University

yili.jin@mila.quebec

Google Scholar

Ralph Cui

Postdoctorat

yufei.cui@mila.quebec

Apprentissage bidirectionnel pour l’optimisation hors ligne basée sur un modèle

Google Scholar

Zonglun Li

Doctorat - McGill University

zonglun.li@mila.quebec

Billets de blogue

Bidirectional Learning for Offline Model-based Optimization

20 septembre 2023

par

Can Chen

Yingxue Zhang

Xue Liu

Mark Coates

Lire l'article

Publications

Think Before You Act: Decision Transformers with Internal Working Memory

Jikun Kang

Romain Laroche

Xingdi Yuan

Adam Trischler

Jie Fu

Large language model (LLM)-based decision-making agents have shown the ability to generalize across multiple tasks. However, their performan… (voir plus)ce relies on massive data and compute. We argue that this inefficiency stems from the forgetting phenomenon, in which a model memorizes its behaviors in parameters throughout training. As a result, training on a new task may deteriorate the model's performance on previous tasks. In contrast to LLMs' implicit memory mechanism, the human brain utilizes distributed memory storage, which helps manage and organize multiple skills efficiently, mitigating the forgetting phenomenon. Thus inspired, we propose an internal working memory module to store, blend, and retrieve information for different downstream tasks. Evaluation results show that the proposed method improves training efficiency and generalization in both Atari games and meta-world object manipulation tasks. Moreover, we demonstrate that memory fine-tuning further enhances the adaptability of the proposed architecture.

2023-05-24

ArXiv (prépublication)

Eliminating Space Scanning: Fast mmWave Beam Alignment with UWB Radios

Ju Wang

Xi Chen

Gregory Dudek

Due to their large bandwidth and impressive data speed, millimeter-wave (mmWave) radios are expected to play a key role in the 5G and beyond… (voir plus) (e.g., 6G) communication networks. Yet, to release mmWave’s true power, the highly directional mmWave beams need to be aligned perfectly. Most existing beam alignment methods adopt an exhaustive or semi-exhaustive space scanning, which introduces up to seconds of delays. To eliminate the need for complex space scanning, this article presents an Ultra-wideband (UWB)-assisted mmWave communication framework, which leverages the co-located UWB antennas to estimate the best angles for mmWave beam alignment. One major challenge of applying this idea in the real world is the barrier of limited antenna numbers. Commercial-Off-The-Shelf (COTS) devices are usually equipped with only a small number of UWB antennas, which are not enough for the existing algorithms to provide an accurate angle estimation. To solve this challenge, we design a novel Multi-Frequency MUltiple SIgnal Classification (MF-MUSIC) algorithm, which extends the classic MUltiple SIgnal Classification (MUSIC) algorithm to the frequency domain and overcomes the antenna limitation barrier in the spatial domain. Extensive real-world experiments and numerical simulations illustrate the advantage of the proposed MF-MUSIC algorithm. MF-MUSIC uses only three antennas to achieve an accurate angle estimation, which is a mere 0.15° (or a relative difference of 3.6%) different from the state-of-the-art 16-antenna-based angle estimation method.

2023-05-16

ACM Transactions on Sensor Networks (publié)

Embracing Channel Estimation in Multi-Packet Reception of ZigBee

Zhe Wang

Linghe Kong

Guihai Chen

As a low-power and low-cost wireless protocol, the promising ZigBee has been widely used in sensor networks and cyber-physical systems. Sinc… (voir plus)e ZigBee based networks usually adopt tree or cluster topology, the convergecast scenarios are common in which multiple transmitters send packets to one receiver, leading to the severe collision problem. The conventional ZigBee adopts carrier sense multiple access with collisions avoidance to avoid collisions, which introduces additional time/energy overhead. The state-of-the-art methods resolve collisions instead of avoidance, in which mZig decomposes a collision by the collision itself and reZig decodes a collision by comparing with reference waveforms. However, mZig falls into high decoding errors only exploiting the signal amplitudes while reZig incurs high computational complexity for waveform comparison. In this paper, we propose CmZig to embrace channel estimation in multiple-packet reception (MPR) of ZigBee, which effectively improves MPR via lightweight computing used for channel estimation and collision decomposition. First, CmZig enables accurate collision decomposition with low computational complexity, which uses the estimated channel parameters modeling both signal amplitudes and phases. Second, CmZig adopts reference waveform comparison only for collisions without chip-level time offsets, instead of the complex machine learning based method. We implement CmZig on USRP-N210 and establish a six-node testbed. Results show that CmZig achieves a bit error rate in the order of

2023-05-01

IEEE Transactions on Mobile Computing (publié)

Structure-aware protein self-supervised learning

Can Chen

Jingbo Zhou

Fan Wang

Dejing Dou

2023-04-13

Bioinformatics (publié)

Intent-aware Multi-source Contrastive Alignment for Tag-enhanced Recommendation

Haolun Wu

Yingxue Zhang

Chen Ma

Wei Guo

Ruiming Tang

Mark Coates

To offer accurate and diverse recommendation services, recent methods use auxiliary information to foster the learning process of user and i… (voir plus)tem representations. Many state-of-the-art (SOTA) methods fuse different sources of information (user, item, knowledge graph, tags, etc.) into a graph and use Graph Neural Networks (GNNs) to introduce the auxiliary information through the message passing paradigm. In this work, we seek an alternative framework that is light and effective through self-supervised learning across different sources of information, particularly for the commonly accessible item tag information. We use a self-supervision signal to pair users with the auxiliary information (tags) associated with the items they have interacted with before. To achieve the pairing, we create a proxy training task. For a given item, the model predicts which is the correct pairing between the representations obtained from the users that have interacted with this item and the tags assigned to it. This design provides an efficient solution, using the auxiliary information directly to enhance the quality of user and item embeddings. User behavior in recommendation systems is driven by the complex interactions of many factors behind the users’ decision-making processes. To make the pairing process more fine-grained and avoid embedding collapse, we propose a user intent-aware self-supervised pairing process where we split the user embeddings into multiple sub-embedding vectors. Each sub-embedding vector captures a specific user intent via self-supervised alignment with a particular cluster of tags. We integrate our designed framework with various recommendation models, demonstrating its flexibility and compatibility. Through comparison with numerous SOTA methods on seven real-world datasets, we show that our method can achieve better performance while requiring less training time. This indicates the potential of applying our approach on web-scale datasets.

2023-04-03

2023 IEEE 39th International Conference on Data Engineering (ICDE) (publié)

Ternary Quantization: A Survey

Danyang Liu

Inference time, model size, and accuracy are critical for deploying deep neural network models. Numerous research efforts have been made to … (voir plus)compress neural network models with faster inference and higher accuracy. Pruning and quantization are mainstream methods to this end. During model quantization, converting individual float values of layer weights to low-precision ones can substantially reduce the computational overhead and improve the inference speed. Many quantization methods have been studied, for example, vector quantization, low-bit quantization, and binary/ternary quantization. This survey focuses on ternary quantization. We review the evolution of ternary quantization and investigate the relationships among existing ternary quantization methods from the perspective of projection function and optimization methods.

2023-03-02

ArXiv (prépublication)

Design and Implementation of Smooth Renewable Power in Cloud Data Centers

Xinxin Liu

Yu Hua

Ling Yang

Yuanyuan Sun

The renewable power has been widely used in modern cloud data centers, which also produce large electricity bills and the negative impacts o… (voir plus)n environments. However, frequent fluctuation and intermittency of renewable power often cause the challenges in terms of the stability of both electricity grid and data centers, as well as decreasing the utilization of renewable power. Existing schemes fail to alleviate the renewable power fluctuation, which is caused by the essential properties of renewable power. In order to address this problem, we propose an efficient and easy-to-use smooth renewable power-aware scheme, called Smoother, which consists of Flexible Smoothing (FS) and Active Delay (AD). First, in order to smooth the fluctuation of renewable power, FS carries out the optimized charge/discharge operation via computing the minimum variance of the renewable power that is supplied to data centers per interval. Second, AD improves the utilization of renewable power via actively adjusting the execution time of deferrable workloads. Extensive experimental results via examining the traces of real-world data centers demonstrate that Smoother significantly reduces the negative impact of renewable power fluctuations on data centers and improves the utilization of renewable power by 250.88 percent on average. We have released the source codes for public use.

2023-03-01

IEEE Transactions on Cloud Computing (publié)

Learning From FM Communications: Toward Accurate, Efficient, All-Terrain Vehicle Localization

Xi Chen

Qiao Xiang

Linghe Kong

Huisan Xu

Vehicle localization service is a fundamental component of intelligent transportation systems. The widely used satellite navigation systems … (voir plus)perform poorly in urban areas because the lines of sight to satellites are blocked by complex terrain characteristics, e.g., buildings, elevated streets and interchanges. In this paper, we design RadioLoc, a novel system achieving accurate, efficient, all-terrain vehicle localization with two key design points. First, RadioLoc harvests the frequency modulation (FM) signal, which has higher availability than satellite signal in complex terrains, as the signal source for localization. Second, RadioLoc integrates modern machine learning techniques into the processing of FM signals to efficiently learn the accurate vehicle localization in all-terrain environments. We validate the feasibility of FM-based vehicle localization and corresponding challenges and practical issues via field tests (e.g., signal distortion, signal inconsistency and limited in- vehicle radio bandwidth), and develop a series of advanced techniques in RadioLoc to address them, including adaptive batching, frequency sweeping, a novel multipath delay spread filter, a reconstructive PCA denoiser and a tailored FM feature extractor. We then develop a generic, modular localization module in RadioLoc, and design different learning-based 3D position identification algorithms for this module. We implement a prototype of RadioLoc and perform extensive field experiments to evaluate its efficiency and efficacy. Results show that (1) RadioLoc achieves a real-time localization latency of less than 100 milliseconds; (2) RadioLoc achieves a worst-case localization accuracy of 99.6% even in an underground parking lot, and (3) the horizontal error of RadioLoc is only one sixth of a dedicated GPS device even when the vehicle is moving at a high-speed (i.e., 80 km/h) in a complex highway scenario.

2023-02-01

IEEE/ACM Transactions on Networking (publié)

Hyperspherical Quantization: Toward Smaller and More Accurate Models

Dan Liu

Xi Chen

Chen Ma

Model quantization enables the deployment of deep neural networks under resource-constrained devices. Vector quantization aims at reducing t… (voir plus)he model size by indexing model weights with full-precision embeddings, i.e., codewords, while the index needs to be restored to 32-bit during computation. Binary and other low-precision quantization methods can reduce the model size up to 32×, however, at the cost of a considerable accuracy drop. In this paper, we propose an efficient framework for ternary quantization to produce smaller and more accurate compressed models. By integrating hyperspherical learning, pruning and reinitialization, our proposed Hyperspherical Quantization (HQ) method reduces the cosine distance between the full-precision and ternary weights, thus reducing the bias of the straight-through gradient estimator during ternary quantization. Compared with existing work at similar compression levels (~30×, ~40×), our method significantly improves the test accuracy and reduces the model size.

2023-01-02

2023 IEEE/CVF Winter Conference on Applications of Computer Vision (WACV) (publié)

Bayes-MIL: A New Probabilistic Perspective on Attention-based Multiple Instance Learning for Whole Slide Images

Yufei Cui

Ziquan Liu

Xiangyu Liu

Cong Wang

Tei-Wei Kuo

Chun Jason Xue

Antoni Bert Chan

Multiple instance learning (MIL) is a popular weakly-supervised learning model on the whole slide image (WSI) for AI-assisted pathology diag… (voir plus)nosis. The recent advance in attention-based MIL allows the model to find its region-of-interest (ROI) for interpretation by learning the attention weights for image patches of WSI slides. However, we empirically find that the interpretability of some related methods is either untrustworthy as the principle of MIL is violated or unsatisfactory as the high-attention regions are not consistent with experts’ annotations. In this paper, we propose Bayes-MIL to address the problem from a probabilistic perspective. The induced patch-level uncertainty is proposed as a new measure of MIL interpretability, which outperforms previous methods in matching doctors annotations. We design a slide-dependent patch regularizer (SDPR) for the attention, imposing constraints derived from the MIL assumption, on the attention distribution. SDPR explicitly constrains the model to generate correct attention values. The spatial information is further encoded by an approximate convolutional conditional random field (CRF), for better interpretability. Experimental results show Bayes-MIL outperforms the related methods in patch-level and slide-level metrics and provides much better interpretable ROI on several large-scale WSI datasets.

2023-01-01

International Conference on Learning Representations (published)

dblp.uni-trier.de

Bayes-MIL: A New Probabilistic Perspective on Attention-based Multiple Instance Learning for Whole Slide Images

Yufei Cui

Ziquan Liu

Xiangyu Liu

Cong Wang

Tei-Wei Kuo

Chun Jason Xue

Antoni B. Chan

2023-01-01

ICLR (publié)

dblp.uni-trier.de

A Distributed Pricing Strategy for Edge Computation Offloading Optimization in Autonomous Driving

Jie Tang

Weilin Zhu

Xiaoming Li

Shaoshan Liu

The increase of on-vehicle applications has brought explosive computation demands to autonomous vehicles and overwhelmed their limited onboa… (voir plus)rd resources. Edge computing can offload application load and effectively alleviate this problem. However, the introduction of edge computing faces significant challenges, including the considerable amount of resource contention due to the scarcity of edge resources and the competition among edge computing resource providers to earn usersâ€™ services requests. We notice that the problem is not purely technical as solutions for these two problems can become conflicting to each other. In this paper, we propose a distributed pricing strategy to achieve full use of computing resources at the edge and maximize the revenue of service operators, both with guaranteed quality-of-service of on-vehicle applications. More specifically, we first use the multi-leader multi-follower Stackelberg game theory to model the pricing of on-vehicle task offloading under edge computing. Next, we propose a distributed pricing strategy to enable edge servers to adjust their local price distributions so that edge servers can bargain with offloading requesters independently. Experimental results confirm that the proposed distributed pricing strategy can provide more optimized server computing resource utilization while guaranteeing the performance of in-vehicle applications.

2023-01-01

IEEE Network (publié)