Search

Recent technological developments allow the online collection of valuable information that can be efficiently used to optimize decisions "on the fly" and at a low cost. These advances have greatly influenced the decision-making process in various areas of operations management, including pricing, inventory, and retail management. In this thesis we study methodological as well as practical aspects arising in online sequential optimization in the presence of such real-time information streams. On the methodological front, we study aspects of sequential optimization in the presence of temporal changes, such as designing decision making policies that adopt to temporal changes in the underlying environment (that drives performance) when only partial information about this changing environment is available, and quantifying the added complexity in sequential decision making problems when temporal changes are introduced. On the applied front, we study practical aspects associated with a class of online services that focus on creating customized recommendations (e.g., Amazon, Netflix). In particular, we focus on online content recommendations, a new class of online services that allows publishers to direct readers from articles they are currently reading to other web-based content they may be interested in, by means of links attached to said article. In the first part of the thesis we consider a non-stationary variant of a sequential stochastic optimization problem, where the underlying cost functions may change along the horizon. We propose a measure, termed {\it variation budget}, that controls the extent of said change, and study how restrictions on this budget impact achievable performance. As a yardstick to quantify performance in non-stationary settings we propose a regret measure relative to a dynamic oracle benchmark. We identify sharp conditions under which it is possible to achieve long-run-average optimality and more refined performance measures such as rate optimality that fully characterize the complexity of such problems. In doing so, we also establish a strong connection between two rather disparate strands of literature: adversarial online convex optimization; and the more traditional stochastic approximation paradigm (couched in a non-stationary setting). This connection is the key to deriving well performing policies in the latter, by leveraging structure of optimal policies in the former. Finally, tight bounds on the minimax regret allow us to quantify the "price of non-stationarity," which mathematically captures the added complexity embedded in a temporally changing environment versus a stationary one. In the second part of the thesis we consider another core stochastic optimization problem couched in a multi-armed bandit (MAB) setting. We develop a MAB formulation that allows for a broad range of temporal uncertainties in the rewards, characterize the (regret) complexity of this class of MAB problems by establishing a direct link between the extent of allowable reward "variation" and the minimal achievable worst-case regret, and provide an optimal policy that achieves that performance. Similarly to the first part of the thesis, our analysis draws concrete connections between two strands of literature: the adversarial and the stochastic MAB frameworks. The third part of the thesis studies applied optimization aspects arising in online content recommendations, that allow web-based publishers to direct readers from articles they are currently reading to other web-based content. We study the content recommendation problem and its unique dynamic features from both theoretical as well as practical perspectives. Using a large data set of browsing history at major media sites, we develop a representation of content along two key dimensions: clickability, the likelihood to click to an article when it is recommended; and engageability, the likelihood to click from an article when it hosts a recommendation. Based on this representation, we propose a class of user path-focused heuristics, whose purpose is to simultaneously ensure a high instantaneous probability of clicking recommended articles, while also optimizing engagement along the future path. We rigorously quantify the performance of these heuristics and validate their impact through a live experiment. The third part of the thesis is based on a collaboration with a leading provider of content recommendations to online publishers.

The past year has been a challenging one. The Steinhardt Museum only opened to the public in July 2018 for a trial run, became fully operational in September, and was gaining significant momentum, but a mere 8 months later, in March 2019, Covid struck and the world as we knew it was transformed. This caught us in the trajectory of an upwards momentum, prior to stabilizing our funding sources and getting government recognition as a museum and in the midst of team-building work and rapid content development. This is not something we had bargained for. That said, our team rose to the occasion. While unfortunately, we had to part with many great hourly workers in the Public Division of the museum and send them on unpaid leave, the remainder of the team stayed with us, and people continued to work as hard as ever in preparation for the museum re-opening, dealing with the many challenges that a new museum affords. The scientific staff continued its work throughout the pandemic. Our scientists and professionals worked in the collections and in the field under very complicated conditions indeed. The collection ma­nagers continued promoting the collections and taxonomic work and providing in­valuable services to a variety of agencies. The Applied Policy-Relevant Research Division Centers—Open Landscape Institute, HaMaarag, and the National Center for Aquatic Ecology—continued almost business as usual, as did the Entomological Laboratory for Applied Ecology. The museum was closed for a significant part of the year but during the months that it was open, albeit with restrictions, and particularly during the summer months, we were gratified by the level of renewed public interest. During the reporting period, the Museum researchers produced 202 scientific articles, which were published or accepted for publication, and13 books/book chapters.

The past year was a wonderful yet challenging one. After a two-month trial run, the museum opened to the public and with almost 200,000 visitors during this past year we have already become the 7th most visited museum in Israel. These numbers were well above our expectations; clearly Israel needed a natural history museum! The Marketing, Operations, and Education teams were committed to welcome this influx of visitors and to develop a new and innovative array of rich and varied programs: workshops, conferences, science days, tours around the galleries and the zoo and botanical garden, night at the museum, plays, public lectures and cool parties. We worked hard to convey our science and values in museum activities, and were gratified by the very positive response we got from our visitors, and thrilled with developing partnerships, in particular with the Tel Aviv Yaffo municipality. Transferring over 5.5 million specimens to the new museum building was challenging, as was reorganizing the specimens in the new storage facilities. With full dedication of the museum’s Collections Managers and Curators, most of the specimens have already been arranged in the new collection halls, where we can now care for them as we should; but it will still take a significant effort to get all our collections in the long term proper storage. The Collections and Research team is growing in numbers and in scientific strength and has reached organizational maturity. The Groups (Terrestrial Vertebrates, Entomology, Marine & Aquatic, Paleosciences, Herbarium) have matured into full-fledged Museum Sections working in coordination and with team spirit. The museum collections database underwent a significant change in the past few years, rearranged on a new software platform that enabled the creation of a unified database that would shortly be publicly available through the museum’s website. Until that happens, we continued to send data as requested to colleagues from around the world. All this was important because hundreds of scientists continued to use our collections for research in the past year, including many colleagues from abroad and graduate students. As a service-oriented institution, we did our best to provide a state-of-the-art research infrastructure for all. Concurrently, the Collections team members continued to record nature assets in areas of their expertise and to promote their taxonomic research and ability to provide crucial services to academia, conservation agencies, agriculture and health authorities, aviation safety, and other missions. The Open Landscape Institute, HaMaarag and the Israel Center for Aquatic Ecology have moved to the new building and, residing now in close physical proximity on the 5th floor, their scientific and professional cooperation is strengthening both among them and with the collections team and research laboratories. These synergies are vital; we plan for the museum to form a whole that is greater than its parts. Moreover, we find interactions with our colleagues from the Ministry of Environmental Protection, Israel Nature and Parks Authority and Keren Kayemeth LeIsrael (Jewish National Fund) through our joint projects both fruitful and scientifically challenging. The museum hosted several national and international meetings. Dafna Langgut was a member of the organizing committee of the 5th Annual Israeli Conference on Environmental History and organized a session on The Steinhardt Natural History Museum as an Archive to Environmental Studies; Shai Meiri was a member of the organizing committee of Gekkota Mundi II | An International Conference; Amos Belmaker was a member of the organizing committee of the 11th Biennial European Bird Curator Meeting; Bella Galil and Menachem Goren organized a workshop Guidelines for the Definition of Deep-Sea Protected Areas within the frame of the Implementation of the MSFD to the Deep Mediterranean Sea multilateral project; Dorothée Huchon organized a Symposium in Honor of Prof. David Wool 86th Birthday; and Jonathan Belmaker represented the museum on the panel of the 2nd Conference of the Israel Chapter of the Society for Conservation Biology that was held at the Technion in Haifa. The Israel Taxonomy Initiative supported three taxonomic workshops on insects (Hemiptera), arrow worms (Chaetognatha) and Peracarida crustaceans. HaMaarag conducted two full-day seminars: Ashalim Stream Ecosystem Montoring Program and Evrona Nature Reserve Ecosystem Monitoring Program. Team building was an important component of the museum’s work in the past year and would continue to be so. The museum team members come from various university units and other organizations, with different areas of expertise and scientific background. Additionally, dozens of new ushers, cashiers and guides were recruited in the past year, and, of course, we shared our building and treasures with a tremendous number of visitors. Thus, we all have to learn to accommodate each other’s needs in the joint building. The greatest challenge facing humankind in the 21st century is the need to provide for a rapidly growing global population, while protecting ecosystems upon which we are all dependent. As a national research infrastructure, the key mission of our museum is to record nature, to study it, and to share our knowledge and expertise with decision-makers and the general public. We feel that the State of Israel and Tel Aviv University have entrusted us with a great treasure, one that is timely, of huge scientific interest, of great societal value, and of immense public interest. We are indeed privileged.