arXiv subjects

Ping-Chun Hsieh

Publications and source records attributed to Ping-Chun Hsieh.

43 records · Page 3Linked to original sources

arXiv2020

Reward-Biased Maximum Likelihood Estimation for Linear Stochastic Bandits

Modifying the reward-biased maximum likelihood method originally proposed in the adaptive control literature, we propose novel learning algorithms to handle the explore-exploit trade-off in linear bandits problems as well as generalized linear bandits problems. We develop novel index policies that we prove achieve order-optimality, and show that they achieve empirical performance competitive with the state-of-the-art benchmark methods in extensive experiments. The new policies achieve this with low computation time per pull for linear bandits, and thereby resulting in both favorable regret as well as computational efficiency.

cs.LG↗

arXiv2020

Developing Multi-Task Recommendations with Long-Term Rewards via Policy Distilled Reinforcement Learning

With the explosive growth of online products and content, recommendation techniques have been considered as an effective tool to overcome information overload, improve user experience, and boost business revenue. In recent years, we have observed a new desideratum of considering long-term rewards of multiple related recommendation tasks simultaneously. The consideration of long-term rewards is strongly tied to business revenue and growth. Learning multiple tasks simultaneously could generally improve the performance of individual task due to knowledge sharing in multi-task learning. While a few existing works have studied long-term rewards in recommendations, they mainly focus on a single recommendation task. In this paper, we propose {\it PoDiRe}: a \underline{po}licy \underline{di}stilled \underline{re}commender that can address long-term rewards of recommendations and simultaneously handle multiple recommendation tasks. This novel recommendation solution is based on a marriage of deep reinforcement learning and knowledge distillation techniques, which is able to establish knowledge sharing among different tasks and reduce the size of a learning model. The resulting model is expected to attain better performance and lower response latency for real-time recommendation services. In collaboration with Samsung Game Launcher, one of the world's largest commercial mobile game platforms, we conduct a comprehensive experimental study on large-scale real data with hundreds of millions of events and show that our solution outperforms many state-of-the-art methods in terms of several standard evaluation metrics.

cs.LG↗

arXiv2019

Stay With Me: Lifetime Maximization Through Heteroscedastic Linear Bandits With Reneging

Sequential decision making for lifetime maximization is a critical problem in many real-world applications, such as medical treatment and portfolio selection. In these applications, a "reneging" phenomenon, where participants may disengage from future interactions after observing an unsatisfiable outcome, is rather prevalent. To address the above issue, this paper proposes a model of heteroscedastic linear bandits with reneging, which allows each participant to have a distinct "satisfaction level," with any interaction outcome falling short of that level resulting in that participant reneging. Moreover, it allows the variance of the outcome to be context-dependent. Based on this model, we develop a UCB-type policy, namely HR-UCB, and prove that it achieves $\mathcal{O}\big(\sqrt{{T}(\log({T}))^{3}}\big)$ regret. Finally, we validate the performance of HR-UCB via simulations.

cs.LG↗

arXiv2018

Streaming Network Embedding through Local Actions

Recently, considerable research attention has been paid to network embedding, a popular approach to construct feature vectors of vertices. Due to the curse of dimensionality and sparsity in graphical datasets, this approach has become indispensable for machine learning tasks over large networks. The majority of existing literature has considered this technique under the assumption that the network is static. However, networks in many applications, nodes and edges accrue to a growing network as a streaming. A small number of very recent results have addressed the problem of embedding for dynamic networks. However, they either rely on knowledge of vertex attributes, suffer high-time complexity or need to be re-trained without closed-form expression. Thus the approach of adapting the existing methods to the streaming environment faces non-trivial technical challenges. These challenges motivate developing new approaches to the problems of streaming network embedding. In this paper, We propose a new framework that is able to generate latent features for new vertices with high efficiency and low complexity under specified iteration rounds. We formulate a constrained optimization problem for the modification of the representation resulting from a stream arrival. We show this problem has no closed-form solution and instead develop an online approximation solution. Our solution follows three steps: (1) identify vertices affected by new vertices, (2) generate latent features for new vertices, and (3) update the latent features of the most affected vertices. The generated representations are provably feasible and not far from the optimal ones in terms of expectation. Multi-class classification and clustering on five real-world networks demonstrate that our model can efficiently update vertex representations and simultaneously achieve comparable or even better performance.

cs.LG↗

arXiv2018

Safe Intersection Management for Mixed Transportation Systems with Human-Driven and Autonomous Vehicles

Most recent studies on establishing intersection safety focus on the situation where all vehicles are fully autonomous. However, currently most vehicles are human-driven and so we will need to transition through regimes featuring a varying proportion of human-driven vehicles ranging from 100\% to 0\% before realizing such a fully autonomous future -- if ever. We will, therefore, need to address the safety of hybrid systems featuring an arbitrary mixture of human-driven and autonomous vehicles. In fact, recent incidents involving autonomous vehicles have already highlighted the need to study the safety of autonomous vehicles co-existing with human-driven vehicles. Motivated by this we address the design of provably safe intersection management for mixed traffic consisting of a mix of human-driven vehicles (HVs) as well as autonomous vehicles (AVs). To analyze such mixed traffic, we model HVs as nearsighted and with relatively loose constraints, permitting worst-case behavior while AVs are considered as capable of following much tighter constraints. HVs are allowed freedom to change their speed at any time while AVs are only allowed to change their speed beginning of a time slot through a Model Predictive Controller (MPC). AVs are assumed to possess a shorter response time and stronger braking capability than HVs in collision avoidance. Moreover, AVs obtain the permissions of passing through the intersection through vehicle-to-infrastructure (V2I) communication, while HVs achieve the same objective by following traffic lights. Taking the above differences into consideration, we propose a provably safe intersection management for mixed traffic comprised of an MPC-based protocol for AVs along with a coordination protocol for traffic lights...

eess.SY↗

arXiv2017

Throughput-Optimal Scheduling for Multi-Hop Networked Transportation Systems With Switch-Over Delay

The emerging connected-vehicle technology provides a new dimension in developing more intelligent traffic control algorithms for signalized intersections in networked transportation systems. An important challenge for the scheduling problem in networked transportation systems is the switch-over delay caused by the guard time before any traffic signal change. The switch-over delay can result in significant loss of system capacity and hence needs to be accommodated in the scheduling design. To tackle this challenge, we propose a distributed online scheduling policy that extends the well-known Max-Pressure policy to address switch-over delay by introducing a bias factor toward the current schedule. We prove that the proposed policy is throughput-optimal with switch-over delay. Furthermore, the proposed policy remains optimal when there are both connected signalized intersections and conventional fixed-time ones in the system. With connected-vehicle technology, the proposed policy can be easily incorporated into the current transportation systems without additional infrastructure. Through extensive simulation in VISSIM, we show that our policy indeed outperforms the existing popular policies.

cs.NI↗

arXiv2017

Delay-Optimal Scheduling for Queueing Systems with Switching Overhead

We study the scheduling polices for asymptotically optimal delay in queueing systems with switching overhead. Such systems consist of a single server that serves multiple queues, and some capacity is lost whenever the server switches to serve a different set of queues. The capacity loss due to this switching overhead can be significant in many emerging applications, and needs to be explicitly addressed in the design of scheduling policies. For example, in 60GHz wireless networks with directional antennas, base stations need to train and reconfigure their beam patterns whenever they switch from one client to another. Considerable switching overhead can also be observed in many other queueing systems such as transportation networks and manufacturing systems. While the celebrated Max-Weight policy achieves asymptotically optimal average delay for systems without switching overhead, it fails to preserve throughput-optimality, let alone delay-optimality, when switching overhead is taken into account. We propose a class of Biased Max-Weight scheduling policies that explicitly takes switching overhead into account. The Biased Max-Weight policy can use either queue length or head-of-line waiting time as an indicator of the system status. We prove that our policies not only are throughput-optimal, but also can be made arbitrarily close to the asymptotic lower bound on average delay. To validate the performance of the proposed policies, we provide extensive simulation with various system topologies and different traffic patterns. We show that the proposed policies indeed achieve much better delay performance than that of the state-of-the-art policy.

cs.PF↗