ترغب بنشر مسار تعليمي؟ اضغط هنا

Deep Learning for Two-Sided Matching

328   0   0.0 ( 0 )
 نشر من قبل Sai Srivatsa Ravindranath
 تاريخ النشر 2021
  مجال البحث الهندسة المعلوماتية
والبحث باللغة English




اسأل ChatGPT حول البحث

We initiate the use of a multi-layer neural network to model two-sided matching and to explore the design space between strategy-proofness and stability. It is well known that both properties cannot be achieved simultaneously but the efficient frontier in this design space is not understood. We show empirically that it is possible to achieve a good compromise between stability and strategy-proofness-substantially better than that achievable through a convex combination of deferred acceptance (stable and strategy-proof for only one side of the market) and randomized serial dictatorship (strategy-proof but not stable).



قيم البحث

اقرأ أيضاً

Two-sided matching platforms provide users with menus of match recommendations. To maximize the number of realized matches between the two sides (referred here as customers and suppliers), the platform must balance the inherent tension between recomm ending customers more potential suppliers to match with and avoiding potential collisions. We introduce a stylized model to study the above trade-off. The platform offers each customer a menu of suppliers, and customers choose, simultaneously and independently, either a supplier from their menu or to remain unmatched. Suppliers then see the set of customers that have selected them, and choose to either match with one of these customers or to remain unmatched. A match occurs if a customer and a supplier choose each other (in sequence). Agents choices are probabilistic, and proportional to public scores of agents in their menu and a score that is associated with remaining unmatched. The platforms problem is to construct menus for costumers to maximize the number of matches. This problem is shown to be strongly NP-hard via a reduction from 3-partition. We provide an efficient algorithm that achieves a constant-factor approximation to the expected number of matches.
We study variants of the stable marriage and college admissions models in which the agents are allowed to express weak preferences over the set of agents on the other side of the market and the option of remaining unmatched. For the problems that we address, previous authors have presented polynomial-time algorithms for computing a Pareto-stable matching. In the case of college admissions, these algorithms require the preferences of the colleges over groups of students to satisfy a technical condition related to responsiveness. We design new polynomial-time Pareto-stable algorithms for stable marriage and college admissions that correspond to strategyproof mechanisms. For stable marriage, it is known that no Pareto-stable mechanism is strategyproof for all of the agents; our algorithm provides a mechanism that is strategyproof for the agents on one side of the market. For college admissions, it is known that no Pareto-stable mechanism can be strategyproof for the colleges; our algorithm provides a mechanism that is strategyproof for the students.
This paper is an attempt to deal with the recent realization (Vazirani, Yannakakis 2021) that the Hylland-Zeckhauser mechanism, which has remained a classic in economics for one-sided matching markets, is likely to be highly intractable. HZ uses the power of a pricing mechanism, which has endowed it with nice game-theoretic properties. Hosseini and Vazirani (2021) define a rich collection of Nash-bargaining-based models for one-sided and two-sided matching markets, in both Fisher and Arrow-Debreu settings, together with implementations using available solvers, and very encouraging experimental results. This naturally raises the question of finding efficient combinatorial algorithms for these models. In this paper, we give efficient combinatorial algorithms based on the techniques of multiplicative weights update (MWU) and conditional gradient descent (CGD) for several one-sided and two-sided models defined in HV 2021. Additionally, we define for the first time a Nash-bargaining-based model for non-bipartite matching markets and solve it using CGD. Furthermore, in every case, we study not only the Fisher but also the Arrow-Debreu version; the latter is also called the exchange version. We give natural applications for each model studied. These models inherit the game-theoretic and computational properties of Nash bargaining. We also establish a deep connection between HZ and the Nash-bargaining-based models, thereby confirming that the alternative to HZ proposed in HV 2021 is a principled one.
Very often in some censorious healthcare scenario, there may be a need to have some expert consultancies (especially by doctors) that are not available in-house to the hospital. With the advancement in technologies (such as video conferencing, smartp hone, etc.), it has become reality that, for the critical medical cases in the hospitals, expert consultants (ECs) from around the world could be hired, who will serve the patients by their physical or virtual presence. Earlier, this interesting healthcare scenario of hiring the ECs (mainly doctors) from outside of the hospitals had been studied with the robust concepts of mechanism design with or without money. We have tried to model the ECs (mainly doctors) hiring problem as a two-sided matching problem. In this paper, for the first time, to the best of our knowledge, we explore the more realistic two-sided matching in our set-up, where the members of the two participating communities, namely patients and doctors are revealing the strict preference ordering over all the members of the opposite community for a stipulated amount of time. We assume that patients and doctors are strategic in nature. With the theoretical analysis, we demonstrate that the proposed mechanism that results in a stable allocation of doctors to patients is strategy-proof (or truthful) and optimal. The proposed mechanism is also validated with exhaustive experiments.
The combination of deep reinforcement learning and search at both training and test time is a powerful paradigm that has led to a number of successes in single-agent settings and perfect-information games, best exemplified by AlphaZero. However, prio r algorithms of this form cannot cope with imperfect-information games. This paper presents ReBeL, a general framework for self-play reinforcement learning and search that provably converges to a Nash equilibrium in any two-player zero-sum game. In the simpler setting of perfect-information games, ReBeL reduces to an algorithm similar to AlphaZero. Results in two different imperfect-information games show ReBeL converges to an approximate Nash equilibrium. We also show ReBeL achieves superhuman performance in heads-up no-limit Texas holdem poker, while using far less domain knowledge than any prior poker AI.

الأسئلة المقترحة

التعليقات
جاري جلب التعليقات جاري جلب التعليقات
سجل دخول لتتمكن من متابعة معايير البحث التي قمت باختيارها
mircosoft-partner

هل ترغب بارسال اشعارات عن اخر التحديثات في شمرا-اكاديميا