BEGIN:VCALENDAR
VERSION:2.0
METHOD:PUBLISH
X-WR-CALNAME: Calendario Seminari | Department of Business and Management: Teaching
PRODID:-//Drupal iCal API//EN
BEGIN:VEVENT
UID:calendar.6487.field_time.0.2
SUMMARY:Where Graphs and Hypergraphs meet Pattern Recognition
DTSTAMP:20210124T193916Z
DTSTART:20210128T140000Z
DTEND:20210128T150000Z
URL;VALUE=URI:https://impresaemanagement.luiss.it/en/seminar/2021/01/21/where-graphs-and-hypergraphs-meet-pattern-recognition
LOCATION:Luiss Research Seminars
DESCRIPTION:Graphs are powerful mathematical entities able to capture topological and \n semantic information from the data at hand. Not by chance they have been \n widely used to model several complex systems in a plethora of domains \n (including\, but not limited to\, biology\, social networks\, computer vision). \n The widespread use of such fascinating structures intrigued machine learning \n engineers and computer scientists alike for more than two decades. Yet \n graphs\, by definition\, are able to capture only pairwise relationships \n amongst vertices and this can limit their modelling power. Hypergraphs fill \n this gap by allowing hyperedges to connect simultaneously two or more \n vertices together. In this talk\, I will be presenting my latest research in \n the development of advanced pattern recognition techniques in the graph \n domain and in the hypergraph domain for solving mainly supervised learning \n problems. The presented techniques span several approaches\, including kernel \n methods\, embedding techniques and feature engineering. In order to \n demonstrate their effectiveness\, two real-world biological case studies will \n also be discussed\, alongside common benchmark tests.
END:VEVENT
BEGIN:VEVENT
UID:calendar.6488.field_time.0.3
SUMMARY:Effective decomposition techniques for hard combinatorial optimization \n problems
DTSTAMP:20210124T193916Z
DTSTART:20210129T100000Z
DTEND:20210129T110000Z
URL;VALUE=URI:https://impresaemanagement.luiss.it/en/seminar/2021/01/21/effective-decomposition-techniques-hard-combinatorial-optimization-problems
LOCATION:Luiss Research Seminars
DESCRIPTION:Divide and conquer\, from Latin divide et impera\, is one of the key techniques \n for tackling combinatorial optimization problems. It relies on the idea of \n decomposing complex problems into a sequence of subproblems that are then \n easier to handle. Decomposition techniques (such as Dantzig-Wolfe\, \n Lagrangian\, or Benders decomposition) are extremely effective in a wide range \n of applications\, including cutting & packing\, production & scheduling\, \n routing & logistics\, telecommunications\, transportation and many others. \n Moreover\, decomposition techniques are playing an important role in many \n different fields of mixed-integer linear and non-linear optimization\, multi \n objective optimization\, optimization under uncertainty\, bilevel optimization\, \n etc. Despite the tremendous amount of research on these topics\, the \n mathematical optimization community is constantly faced with new challenges \n coming from theoretical aspects and real world applications that require the \n development of new advanced tools. In this talk\, I will present \n state-of-the-art decomposition techniques for several combinatorial \n optimization problems with a particular attention in covering\, location\, \n coloring and general mixed integer linear programs.
END:VEVENT
BEGIN:VEVENT
UID:calendar.6489.field_time.0.4
SUMMARY:Distributed Algorithms for Large-Scale Graphs and Networks
DTSTAMP:20210124T193916Z
DTSTART:20210129T143000Z
DTEND:20210129T153000Z
URL;VALUE=URI:https://impresaemanagement.luiss.it/en/seminar/2021/01/21/distributed-algorithms-large-scale-graphs-and-networks
LOCATION:Luiss Research Seminars
DESCRIPTION:As massive graphs become more prevalent\, there is a rapidly growing need \n for scalable distributed algorithms that solve fundamental graph problems on \n very large datasets. At the same time\, large distributed networks such as the \n Internet\, sensor networks\, ad hoc wireless networks\, and the IoT are becoming \n ubiquitous\, increasing the need for scalable network algorithms that provide \n basic primitives. In this talk I will discuss some recent developments in \n these two contexts\, highlighting the commonalities between them.
END:VEVENT
BEGIN:VEVENT
UID:calendar.6490.field_time.0.5
SUMMARY:Consumers believe that products work better for others
DTSTAMP:20210124T193916Z
DTSTART:20210129T163000Z
DTEND:20210129T180000Z
URL;VALUE=URI:https://impresaemanagement.luiss.it/en/seminar/2021/01/21/consumers-believe-products-work-better-others
LOCATION:Luiss Research Seminars
DESCRIPTION:Hundreds of studies have shown that consumers tend to see themselves in the \n best possible light\, yet we present evidence that consumers have a \n surprisingly glum perspective on receiving a product’s claimed effects. In \n 12 studies (N = 5\,855\; including 9 pre-registered)\, we found that consumers \n believe that product efficacy is higher for others than it is for themselves. \n For example\, consumers believe that consuming products like an adult coloring \n book (to inspire creativity)\, or a sports drink (to satisfy thirst)\, or \n medicine (to relieve pain)\, or an online class (to learn something new) will \n have a greater effect on others than on themselves. We show that this bias \n holds across many kinds of products and judgment-targets\, and inversely \n correlates with factors such as: product-familiarity\, product-usefulness\, and \n relationship closeness with judgment-targets. We evidence that this bias \n stems from the fact that consumers believe they are more unique than others\, \n and less malleable\; and we show that this bias in perceived product efficacy \n alters the choices that consumers make for others. We conclude by discussing \n implications for research on gift-giving\, advice-giving\, usership\, and for \n interpersonal social-\, health-\, and financial-choices.
END:VEVENT
BEGIN:VEVENT
UID:calendar.6466.field_time.0.6
SUMMARY:TBD
DTSTAMP:20210124T193916Z
DTSTART:20210218T110000Z
DTEND:20210218T120000Z
URL;VALUE=URI:https://impresaemanagement.luiss.it/en/seminar/2020/12/10/tbd-0
LOCATION:Luiss Research Seminars
DESCRIPTION:TBD
END:VEVENT
BEGIN:VEVENT
UID:calendar.6467.field_time.0.7
SUMMARY:TBD
DTSTAMP:20210124T193916Z
DTSTART:20210225T110000Z
DTEND:20210225T120000Z
URL;VALUE=URI:https://impresaemanagement.luiss.it/en/seminar/2020/12/10/tbd-1
LOCATION:Luiss Research Seminars
DESCRIPTION:TBD
END:VEVENT
END:VCALENDAR