Mix and Match: Markov Chains and Mixing Times for Matching in Rideshare

Rideshare platforms such as Uber and Lyft dynamically dispatch drivers to match riders' requests. We model the dispatching process in rideshare as a Markov chain that takes into account the geographic mobility of both drivers and riders over time. Prior work explores dispatch policies in the limit of such Markov chains; we characterize when this limit assumption is valid, under a variety of natural dispatch policies. We give explicit bounds on convergence in general, and exact (including constants) convergence rates for special cases. Then, on simulated and real transit data, we show that our bounds characterize convergence rates -- even when the necessary theoretical assumptions are relaxed. Additionally these policies compare well against a standard reinforcement learning algorithm which optimizes for profit without any convergence properties.

[1]  W. K. Hastings,et al.  Monte Carlo Sampling Methods Using Markov Chains and Their Applications , 1970 .

[2]  Shane Legg,et al.  Human-level control through deep reinforcement learning , 2015, Nature.

[3]  Patrick Jaillet,et al.  Online Spatio-Temporal Matching in Stochastic and Dynamic Domains , 2016, AAAI.

[4]  Wei Xu,et al.  An Optimization Framework for Online Ride-Sharing Markets , 2016, 2017 IEEE 37th International Conference on Distributed Computing Systems (ICDCS).

[5]  Lei Ying,et al.  Empty-Car Routing in Ridesharing Systems , 2016, Oper. Res..

[6]  Jun Wang,et al.  Online Algorithms of Task Allocation in Spatial Crowdsourcing , 2017, ChineseCSCW.

[7]  Emilio Frazzoli,et al.  On-demand high-capacity ride-sharing via dynamic trip-vehicle assignment , 2017, Proceedings of the National Academy of Sciences.

[8]  Zhe Xu,et al.  Efficient Large-Scale Fleet Management via Multi-Agent Deep Reinforcement Learning , 2018, KDD.

[9]  Aranyak Mehta,et al.  Online Matching and Ad Allocation , 2013, Found. Trends Theor. Comput. Sci..

[10]  Nando de Freitas,et al.  An Introduction to MCMC for Machine Learning , 2004, Machine Learning.

[11]  Patrick Jaillet,et al.  Rebalancing shared mobility-on-demand systems: A reinforcement learning approach , 2017, 2017 IEEE 20th International Conference on Intelligent Transportation Systems (ITSC).

[12]  Carlos Riquelme,et al.  Dynamic pricing in ridesharing platforms , 2016, SECO.

[13]  Yuxi Li,et al.  Deep Reinforcement Learning: An Overview , 2017, ArXiv.

[14]  Siddhartha Banerjee,et al.  State Dependent Control of Closed Queueing Networks with Application to Ride-Hailing , 2018 .

[15]  Venkatesan Guruswami.  Rapidly Mixing Markov Chains: A Comparison of Techniques (A Survey) , 2016, ArXiv.

[16]  Thodoris Lykouris,et al.  Pricing and Optimization in Shared Vehicle Systems: An Approximation Framework , 2016, EC.

[17]  Patrick Jaillet,et al.  Dynamic Repositioning to Reduce Lost Demand in Bike Sharing Systems , 2017, J. Artif. Intell. Res..

[18]  V. Climenhaga.  Markov chains and mixing times , 2013 .

[19]  Robert Hahn,et al.  The Ridesharing Revolution : Economic Survey and Synthesis , 2017 .

[20]  J. Dickerson,et al.  FutureMatch: Learning to Match in Dynamic Environments.: Abstract# A2987 , 2014 .

[21]  Tuomas Sandholm,et al.  FutureMatch: Combining Human Value Judgments and Machine Learning to Match in Dynamic Environments , 2015, AAAI.

[22]  David B. Shmoys,et al.  Data Analysis and Optimization for (Citi)Bike Sharing , 2015, AAAI.

[23]  David C. Parkes,et al.  Spatio-Temporal Pricing for Ridesharing Platforms , 2018, EC.

[24]  Gérard P. Cachon,et al.  The Role of Surge Pricing on a Service Platform with Self-Scheduling Capacity , 2016, Manuf. Serv. Oper. Manag..

[25]  A. Rollett,et al.  The Monte Carlo Method , 2004 .

[26]  J. Yosinski,et al.  Time-series Extreme Event Forecasting with Neural Networks at Uber , 2017 .

[27]  Amy R. Ward,et al.  Dynamic Matching for Real-Time Ride Sharing , 2020, Stochastic Systems.

[28]  Y. Peres,et al.  Markov Chains and Mixing Times: Second Edition , 2017 .

[29]  Richard S. Sutton,et al.  Reinforcement Learning: An Introduction , 1998, IEEE Trans. Neural Networks.

[30]  Guannan Liu,et al.  A cost-effective recommender system for taxi drivers , 2014, KDD.

[31]  Lei Chen,et al.  Online Minimum Matching in Real-Time Spatial Data: Experiments and Analysis , 2016, Proc. VLDB Endow..

[32]  Jieping Ye,et al.  Deep Multi-View Spatial-Temporal Network for Taxi Demand Prediction , 2018, AAAI.

[33]  Tao Qin,et al.  Dynamic Pricing for Reusable Resources in Competitive Market With Stochastic Demand , 2018, AAAI.

[34]  Aravind Srinivasan,et al.  Allocation Problems in Ride-sharing Platforms , 2017, AAAI.

[35]  N. Metropolis,et al.  Equation of State Calculations by Fast Computing Machines , 1953, Resonance.

[36]  Sarit Kraus,et al.  Augmenting Decisions of Taxi Drivers through Reinforcement Learning for Improving Revenues , 2017, ICAPS.

[37]  Chaitanya Swamy,et al.  Minimizing Latency in Online Ride and Delivery Services , 2018, WWW.