Related papers: A Better Match for Drivers and Riders: Reinforcement Learning at Lyft

A Better Match for Drivers and Riders: Reinforcement Learning at Lyft

URL: http://arxiv.org/abs/2310.13810v2
Date: Tue, 14 Nov 2023 02:30:34 GMT
Title: A Better Match for Drivers and Riders: Reinforcement Learning at Lyft
Authors: Xabi Azagirre, Akshay Balwally, Guillaume Candeli, Nicholas Chamandy, Benjamin Han, Alona King, Hyungjun Lee, Martin Loncaric, Sebastien Martin, Vijay Narasiman, Zhiwei (Tony) Qin, Baptiste Richard, Sara Smoot, Sean Taylor, Garrett van Ryzin, Di Wu, Fei Yu, Alex Zamoshchin
Abstract summary: We use a novel online reinforcement learning approach that estimates the future earnings of drivers in real time. This change was the first documented implementation of a ridesharing matching algorithm that can learn and improve in real time. Lyft rolled out the algorithm globally in 2021.
Score: 9.901159075969318
License: http://creativecommons.org/licenses/by/4.0/
Abstract: To better match drivers to riders in our ridesharing application, we revised Lyft's core matching algorithm. We use a novel online reinforcement learning approach that estimates the future earnings of drivers in real time and use this information to find more efficient matches. This change was the first documented implementation of a ridesharing matching algorithm that can learn and improve in real time. We evaluated the new approach during weeks of switchback experimentation in most Lyft markets, and estimated how it benefited drivers, riders, and the platform. In particular, it enabled our drivers to serve millions of additional riders each year, leading to more than $30 million per year in incremental revenue. Lyft rolled out the algorithm globally in 2021.

Related papers

Not Even Nice Work If You Can Get It; A Longitudinal Study of Uber's Algorithmic Pay and Pricing [20.694356269172857]
Ride-sharing platforms like Uber market themselves as enabling flexibility' for their workforce.<n>We describe our process of participatory action research with drivers and trade union organisers, culminating in a participatory audit of Uber's algorithmic pay and work allocation.<n>We find that after dynamic pricing, pay has decreased, Uber's cut has increased, job allocation and pay is less predictable, inequality between drivers is increased, and drivers spend more time waiting for jobs.
arXiv Detail & Related papers (2025-06-18T08:59:39Z)
Nearly-Optimal Bandit Learning in Stackelberg Games with Side Information [57.287431079644705]
We study the problem of online learning in Stackelberg games with side information between a leader and a sequence of followers. We provide learning algorithms for the leader which achieve $O(T1/2)$ regret under bandit feedback.
arXiv Detail & Related papers (2025-01-31T22:40:57Z)
User Strategization and Trustworthy Algorithms [81.82279667028423]
We show that user strategization can actually help platforms in the short term. We then show that it corrupts platforms' data and ultimately hurts their ability to make counterfactual decisions.
arXiv Detail & Related papers (2023-12-29T16:09:42Z)
Race Driver Evaluation at a Driving Simulator using a physical Model and a Machine Learning Approach [1.9395755884693817]
We present a method to study and evaluate race drivers on a driver-in-the-loop simulator. An overall performance score, a vehicle-trajectory score and a handling score are introduced to evaluate drivers. We show that the neural network is accurate and robust with a root-mean-square error between 2-5% and can replace the optimisation based method.
arXiv Detail & Related papers (2022-01-27T07:32:32Z)
DriverGym: Democratising Reinforcement Learning for Autonomous Driving [75.91049219123899]
We propose DriverGym, an open-source environment for developing reinforcement learning algorithms for autonomous driving. DriverGym provides access to more than 1000 hours of expert logged data and also supports reactive and data-driven agent behavior. The performance of an RL policy can be easily validated on real-world data using our extensive and flexible closed-loop evaluation protocol.
arXiv Detail & Related papers (2021-11-12T11:47:08Z)
Value Function is All You Need: A Unified Learning Framework for Ride Hailing Platforms [57.21078336887961]
Large ride-hailing platforms, such as DiDi, Uber and Lyft, connect tens of thousands of vehicles in a city to millions of ride demands throughout the day. We propose a unified value-based dynamic learning framework (V1D3) for tackling both tasks.
arXiv Detail & Related papers (2021-05-18T19:22:24Z)
Real-world Ride-hailing Vehicle Repositioning using Deep Reinforcement Learning [52.2663102239029]
We present a new practical framework based on deep reinforcement learning and decision-time planning for real-world vehicle on idle-hailing platforms. Our approach learns ride-based state-value function using a batch training algorithm with deep value. We benchmark our algorithm with baselines in a ride-hailing simulation environment to demonstrate its superiority in improving income efficiency.
arXiv Detail & Related papers (2021-03-08T05:34:05Z)
Driver2vec: Driver Identification from Automotive Data [44.84876493736275]
Driver2vec is able to accurately identify the driver from a short 10-second interval of sensor data. Driver2vec is trained on a dataset of 51 drivers provided by Nervtech.
arXiv Detail & Related papers (2021-02-10T03:09:13Z)
A multi-objective optimization framework for on-line ridesharing systems [6.247570729758392]
We propose an algorithm that leverages biogeography-based optimization to solve a multi-objective optimization problem for online ridesharing. We test our algorithm by evaluatingperformance on the Beijing ridesharing dataset.
arXiv Detail & Related papers (2020-12-07T16:25:39Z)
Predicting Requests in Large-Scale Online P2P Ridesharing [1.8434430658837255]
Peer-to-peer ridesharing (P2P-RS) enables people to arrange one-time rides with their own private cars, without the involvement of professional drivers. In this paper we tackle the question of assessing the benefit of predicting ridesharing requests in the context of P2P-RS optimisation. Results on a public real-world show that, by employing a perfect predictor, the total reward can be improved by 5.27% with a forecast horizon of 1 minute.
arXiv Detail & Related papers (2020-09-07T10:27:24Z)
Balancing Taxi Distribution in A City-Scale Dynamic Ridesharing Service: A Hybrid Solution Based on Demand Learning [0.0]
We study the challenging problem of how to balance taxi distribution across a city in a dynamic ridesharing service. We propose a hybrid solution involving a series of algorithms: the Correlated Pooling collects correlated rider requests, the Adjacency Ride-Matching based on Demand Learning assigns taxis to riders, and the Greedy Idle Movement aims to direct taxis without a current assignment to the areas with riders in need of service.
arXiv Detail & Related papers (2020-07-27T07:08:02Z)
Cost-sensitive Multi-class AdaBoost for Understanding Driving Behavior with Telematics [1.4502611532302039]
Insurers can now capture a wide range of data, such as distance traveled, how drivers brake, accelerate or make turns, and travel frequency each day of the week. Such additional information helps insurers improve risk assessments for usage-based insurance (UBI) In this article, we explore how to integrate telematics information to better predict claims frequency.
arXiv Detail & Related papers (2020-07-06T22:26:56Z)
Road Network Metric Learning for Estimated Time of Arrival [93.0759529610483]
In this paper, we propose the Road Network Metric Learning framework for Estimated Time of Arrival (ETA) It consists of two components: (1) a main regression task to predict the travel time, and (2) an auxiliary metric learning task to improve the quality of link embedding vectors. We show that our method outperforms the state-of-the-art model and the promotion concentrates on the cold links with few data.
arXiv Detail & Related papers (2020-06-24T04:45:14Z)

This list is automatically generated from the titles and abstracts of the papers in this site.