We study large-scale machine learning problems in changing environments where a small part of the dataset is modified, and the effect of the data modification must be monitored in order to know how much the modification changes the optimal model. When the entire dataset is large, even if the amount of the data modification is fairly small, the computational cost for re-training the model would be prohibitively large. In this paper, we propose a novel method, called the optimal solution bounding (OSB), for monitoring such a data modification effect on the optimal model by efficiently evaluating (without actually re-training) it. The proposed method provides bounds on the unknown optimal model with the cost proportional only to the size of the data modification.

In a seminal paper, McAfee (1992) presented a truthful mechanism for double auctions, attaining asymptotically-optimal gain-from-trade without any prior information on the valuations of the traders. McAfee's mechanism handles single-parametric agents, allowing each seller to sell a single unit and each buyer to buy a single unit. This paper presents a double-auction mechanism that handles multi-parametric agents and allows multiple units per trader, as long as the valuation functions of all traders have decreasing marginal returns. The mechanism is prior-free, ex-post individually-rational, dominant-strategy truthful and strongly-budget-balanced. Its gain-from-trade approaches the optimum when the market size is sufficiently large.

This paper examines two-sided matching with budget constraints where one side (a firm or hospital) can make monetary transfers (offer wages) to the other (a worker or doctor). In a standard model, while multiple doctors can be matched to a single hospital, a hospital has a maximum quota; thus, the number of doctors assigned to a hospital cannot exceed a certain limit. In our model, in contrast, a hospital has a fixed budget; that is, the total amount of wages allocated by each hospital to doctors is constrained. With budget constraints, stable matchings may fail to exist and checking for the existence is hard. To deal with the nonexistence of stable matchings, we extend the "matching with contracts" model of Hatfield and Milgrom so that it deals with approximately stable matchings where each of the hospitals' utilities after deviation can increase by a factor up to a certain amount. We then propose two novel mechanisms that efficiently return a stable matching that exactly satisfies the budget constraints. Specifically, by sacrificing strategy-proofness, our first mechanism achieves the best possible bound. We also explore a special case on which a simple mechanism is strategy-proof for doctors, while maintaining the best possible bound of the general case.

In this paper, we propose a fractional preference model for the facility location game with two facilities that serve the similar purpose on a line where each agent has his location information as well as fractional preference to indicate how well they prefer the facilities. The preference for each facility is in the range of [0, L] such that the sum of the preference for all facilities is equal to 1. The utility is measured by subtracting the sum of the cost of both facilities from the total length L where the cost of facilities is defined as the multiplication of the fractional preference and the distance between the agent and the facilities. We first show that the lower bound for the objective of minimizing total cost is at least Ω(n^1/3). Hence, we use the utility function to analyze the agents' satification. Our objective is to place two facilities on [0, L] to maximize the social utility or the minimum utility. For each objective function, we propose deterministic strategy-proof mechanisms. For the objective of maximizing the social utility, we present an optimal deterministic strategy-proof mechanism in the case where agents can only misreport their locations. In the case where agents can only misreport their preferences, we present a 2-approximation deterministic strategy-proof mechanism. Finally, we present a 4-approximation deterministic strategy-proof mechanism and a randomized strategy-proof mechanism with an approximation ratio of 2 where agents can misreport both the preference and location information. Moreover, we also give a lower-bound of 1.06. For the objective of maximizing the minimum utility, we give a lower-bound of 1.5 and present a 2-approximation deterministic strategy-proof mechanism where agents can misreport both the preference and location.

Treatable infectious diseases are a critical challenge for public health. Outreach campaigns can encourage undiagnosed patients to seek treatment but must be carefully targeted to make the most efficient use of limited resources. We present an algorithm to optimally allocate limited outreach resources among demographic groups in the population. The algorithm uses a novel multiagent model of disease spread which both captures the underlying population dynamics and is amenable to optimization. Our algorithm extends, with provable guarantees, to a stochastic setting where we have only a distribution over parameters such as the contact pattern between agents. We evaluate our algorithm on two instances where this distribution is inferred from real world data: tuberculosis in India and gonorrhea in the United States. Our algorithm produces a policy which is predicted to avert an average of least 8,000 person-years of tuberculosis and 20,000 person-years of gonorrhea annually compared to current policy.

Peer grading, allowing students/peers to evaluate others' assignments, offers a promising solution for scaling evaluation and learning to large-scale educational systems. A key challenge in peer grading is motivating peers to grade diligently. While existing spot-checking (SC) mechanisms can prevent peer collusion where peers coordinate to report the uninformative grade, they unrealistically assume that peers have the same grading reliability and cost. This paper studies the general Optimal Spot-Checking (OptSC) problem of determining the probability each assignment needs to be checked to maximize assignments' evaluation accuracy aggregated from peers, and takes into consideration 1) peers' heterogeneous characteristics, and 2) peers' strategic grading behaviors to maximize their own utility. We prove that the bilevel OptSC is NP-hard to solve. By exploiting peers' grading behaviors, we first formulate a single level relaxation to approximate OptSC. By further exploiting structural properties of the relaxed problem, we propose an efficient algorithm to that relaxation, which also gives a good approximation of the original OptSC. Extensive experiments on both synthetic and real datasets show significant advantages of the proposed algorithm over existing approaches.

The heavy burdens of computation and off-chip traffic impede deploying the large scale convolution neural network on embedded platforms. As CNN is attributed to the strong endurance to computation errors, employing block floating point (BFP) arithmetics in CNN accelerators could save the hardware cost and data traffics efficiently, while maintaining the classification accuracy. In this paper, we verify the effects of word width definitions in BFP to the CNN performance without retraining. Several typical CNN models, including VGG16, ResNet-18, ResNet-50 and GoogLeNet, were tested in this paper. Experiments revealed that 8-bit mantissa, including sign bit, in BFP representation merely induced less than 0.3% accuracy loss. In addition, we investigate the computational errors in theory and develop the noise-to-signal ratio (NSR) upper bound, which provides the promising guidance for BFP based CNN engine design.

Predicting epidemic dynamics is of great value in understanding and controlling diffusion processes, such as infectious disease spread and information propagation. This task is intractable, especially when surveillance resources are very limited. To address the challenge, we study the problem of active surveillance, i.e., how to identify a small portion of system components as sentinels to effect monitoring, such that the epidemic dynamics of an entire system can be readily predicted from the partial data collected by such sentinels. We propose a novel measure, the gamma value, to identify the sentinels by modeling a sentinel network with row sparsity structure. We design a flexible group sparse Bayesian learning algorithm to mine the sentinel network suitable for handling both linear and non-linear dynamical systems by using the expectation maximization method and variational approximation. The efficacy of the proposed algorithm is theoretically analyzed and empirically validated using both synthetic and real-world data.

Big human mobility data are being continuously generated through a variety of sources, some of which can be treated and used as streaming data for understanding and predicting urban dynamics. With such streaming mobility data, the online prediction of short-term human mobility at the city level can be of great significance for transportation scheduling, urban regulation, and emergency management. In particular, when big rare events or disasters happen, such as large earthquakes or severe traffic accidents, people change their behaviors from their routine activities. This means people's movements will almost be uncorrelated with their past movements. Therefore, in this study, we build an online system called DeepUrbanMomentum to conduct the next short-term mobility predictions by using (the limited steps of) currently observed human mobility data. A deep-learning architecture built with recurrent neural networks is designed to effectively model these highly complex sequential data for a huge urban area. Experimental results demonstrate the superior performance of our proposed model as compared to the existing approaches. Lastly, we apply our system to a real emergency scenario and demonstrate that our system is applicable in the real world.

To alleviate traffic congestion in urban areas, electronic toll collection (ETC) systems are deployed all over the world. Despite the merits, tolls are usually pre-determined and fixed from day to day, which fail to consider traffic dynamics and thus have limited regulation effect when traffic conditions are abnormal. In this paper, we propose a novel dynamic ETC (DyETC) scheme which adjusts tolls to traffic conditions in realtime. The DyETC problem is formulated as a Markov decision process (MDP), the solution of which is very challenging due to its 1) multi-dimensional state space, 2) multi-dimensional, continuous and bounded action space, and 3) time-dependent state and action values. Due to the complexity of the formulated MDP, existing methods cannot be applied to our problem. Therefore, we develop a novel algorithm, PG-beta, which makes three improvements to traditional policy gradient method by proposing 1) time-dependent value and policy functions, 2) Beta distribution policy function and 3) state abstraction. Experimental results show that, compared with existing ETC schemes, DyETC increases traffic volume by around 8%, and reduces travel time by around 14:6% during rush hour. Considering the total traffic volume in a traffic network, this contributes to a substantial increase to social welfare.