HEC Montréal

About: HEC Montréal is a education organization based out in Montreal, Quebec, Canada. It is known for research contribution in the topics: Vehicle routing problem & Corporate governance. The organization has 1221 authors who have published 5708 publications receiving 196862 citations. The organization is also known as: Ecole des Hautes Etudes Commerciales de Montreal & HEC Montreal.

InfoGraph: Unsupervised and Semi-supervised Graph-Level Representation Learning via Mutual Information Maximization

Abstract: This paper studies learning the representations of whole graphs in both unsupervised and semi-supervised scenarios. Graph-level representations are critical in a variety of real-world applications such as predicting the properties of molecules and community analysis in social networks. Traditional graph kernel based methods are simple, yet effective for obtaining fixed-length representations for graphs but they suffer from poor generalization due to hand-crafted designs. There are also some recent methods based on language models (e.g. graph2vec) but they tend to only consider certain substructures (e.g. subtrees) as graph representatives. Inspired by recent progress of unsupervised representation learning, in this paper we proposed a novel method called InfoGraph for learning graph-level representations. We maximize the mutual information between the graph-level representation and the representations of substructures of different scales (e.g., nodes, edges, triangles). By doing so, the graph-level representations encode aspects of the data that are shared across different scales of substructures. Furthermore, we further propose InfoGraph*, an extension of InfoGraph for semisupervised scenarios. InfoGraph* maximizes the mutual information between unsupervised graph representations learned by InfoGraph and the representations learned by existing supervised methods. As a result, the supervised encoder learns from unlabeled data while preserving the latent semantic space favored by the current supervised task. Experimental results on the tasks of graph classification and molecular property prediction show that InfoGraph is superior to state-of-the-art baselines and InfoGraph* can achieve performance competitive with state-of-the-art semi-supervised models.

The regional dimension of MNEs' foreign subsidiary localization

Abstract: This paper examines the regional effect of MNEs' foreign subsidiary localization. We hypothesize that the number of subsequent foreign subsidiaries in a country is in part determined by a firm's prior foreign subsidiary activity at the regional level. We test our hypotheses using data on 1076 Japanese MNEs that created 3466 foreign subsidiaries (1837 wholly owned FDIs and 1629 joint ventures) over the period 1996–2001. We use a multilevel negative binomial approach with three levels of analysis: localization decisions in a country (49 countries), in a region (six regions) and at the headquarters level. In this way, we test the regional effects controlling for country and corporate dimensions. We also run separate models to differentiate wholly owned and joint venture localization decisions. Our results strongly support the semi-globalization perspective in that the regional-level effects are significant and different from the country-level effects for all foreign subsidiaries, for wholly owned subsidiaries and for jointly owned subsidiaries. Japanese MNEs adopt a regional perspective that complements their decisions at the country and firm levels. They seek regional agglomeration benefits and make arbitrage decisions between countries in the same region.

Solving a home-care districting problem in an urban setting

Abstract: This article describes a districting study undertaken for the Cote-des-Neiges local community health clinic in Montreal. A territory must be partitioned into six districts by suitably grouping territorial basic units. Five districting criteria must be respected: indivisibility of basic units, respect for borough boundaries, connectivity, visiting personnel mobility, and workload equilibrium. The last two criteria are combined into a single objective function and the problem is solved by means of a tabu search technique that iteratively moves a basic unit to an adjacent district or swaps two basic units between adjacent districts. The problem was solved and the clinic management confirmed its satisfaction after a 2 year implementation period.