arXiv subjects

Qin

Publications and source records attributed to Qin.

4 recordsLinked to original sources

arXiv2023

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

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.

cs.LG↗

arXiv2019

Efficient Ridesharing Order Dispatching with Mean Field Multi-Agent Reinforcement Learning

A fundamental question in any peer-to-peer ridesharing system is how to, both effectively and efficiently, dispatch user's ride requests to the right driver in real time. Traditional rule-based solutions usually work on a simplified problem setting, which requires a sophisticated hand-crafted weight design for either centralized authority control or decentralized multi-agent scheduling systems. Although recent approaches have used reinforcement learning to provide centralized combinatorial optimization algorithms with informative weight values, their single-agent setting can hardly model the complex interactions between drivers and orders. In this paper, we address the order dispatching problem using multi-agent reinforcement learning (MARL), which follows the distributed nature of the peer-to-peer ridesharing problem and possesses the ability to capture the stochastic demand-supply dynamics in large-scale ridesharing scenarios. Being more reliable than centralized approaches, our proposed MARL solutions could also support fully distributed execution through recent advances in the Internet of Vehicles (IoV) and the Vehicle-to-Network (V2N). Furthermore, we adopt the mean field approximation to simplify the local interactions by taking an average action among neighborhoods. The mean field approximation is capable of globally capturing dynamic demand-supply variations by propagating many local interactions between agents and the environment. Our extensive experiments have shown the significant improvements of MARL order dispatching algorithms over several strong baselines on the gross merchandise volume (GMV), and order response rate measures. Besides, the simulated experiments with real data have also justified that our solution can alleviate the supply-demand gap during the rush hours, thus possessing the capability of reducing traffic congestion.

cs.MA↗

arXiv2017

A Unified Neural Network Approach for Estimating Travel Time and Distance for a Taxi Trip

In building intelligent transportation systems such as taxi or rideshare services, accurate prediction of travel time and distance is crucial for customer experience and resource management. Using the NYC taxi dataset, which contains taxi trips data collected from GPS-enabled taxis [23], this paper investigates the use of deep neural networks to jointly predict taxi trip time and distance. We propose a model, called ST-NN (Spatio-Temporal Neural Network), which first predicts the travel distance between an origin and a destination GPS coordinate, then combines this prediction with the time of day to predict the travel time. The beauty of ST-NN is that it uses only the raw trips data without requiring further feature engineering and provides a joint estimate of travel time and distance. We compare the performance of ST-NN to that of state-of-the-art travel time estimation methods, and we observe that the proposed approach generalizes better than state-of-the-art methods. We show that ST-NN approach significantly reduces the mean absolute error for both predicted travel time and distance, about 17% for travel time prediction. We also observe that the proposed approach is more robust to outliers present in the dataset by testing the performance of ST-NN on the datasets with and without outliers.

stat.ML↗

arXiv2012

The molecular emissions and the infall motion in the high-mass young stellar object G8.68-0.37

We present a multi-wavelength observational study towards the high-mass young stellar object G8.68-0.37. A single massive gas-and-dust core is observed in the (sub)millimeter continuum and molecular line emissions. We fitted the spectral energy distribution (SED) from the dust continuum emission. The best-fit SED suggests the presence of two components with temperature of $T_{\rm d}=20$ K and 120 K, respectively. The core has a total mass of up to $1.5\times10^3$ $M_{\odot}$ and bolometric luminosity of $2.3\times10^4 L_{\odot}$. Both the mass and luminosity are dominated by the cold component ($T_{\rm d}=20$ K). The molecular lines of C$^{18}$O, C$^{34}$S, DCN, and thermally excited CH$_3$OH are detected in this core. Prominent infall signatures are observed in the $^{12}$CO $(1-0)$ and $(2-1)$. We estimated an infall velocity of 0.45 km s$^{-1}$ and mass infall rate of $7\times10^{-4} M_{\odot}$ year$^{-1}$. From the molecular lines, we have found a high DCN abundance and relative abundance ratio to HCN. The overabundant DCN may originate from a significant deuteration in the previous cold pre-protostellar phase. And the DCN should now be rapidly sublimated from the grain mantles to maintain the overabundance in the gas phase.

astro-ph.SR↗