Approximating Martingale Process (AMP) is proven to be effective for variance reduction in reinforcement learning (RL) in specific cases such as Multiclass Queueing Networks. However, in the already proven cases, the state space is relatively small and all possible state transitions can be iterated through. In this paper, we consider systems in which state space is large and have uncertainties when considering state transitions, thus making AMP a generalized variance-reduction method in RL. Specifically, we will investigate the application of AMP in ride-hailing systems like Uber, where Proximal Policy Optimization (PPO) is incorporated to optimize the policy of matching drivers and customers.

Recent work has shown the benefits of synthetic data for use in computer vision, with applications ranging from autonomous driving to face landmark detection and reconstruction. There are a number of benefits of using synthetic data from privacy preservation and bias elimination to quality and feasibility of annotation. Generating human-centered synthetic data is a particular challenge in terms of realism and domain-gap, though recent work has shown that effective machine learning models can be trained using synthetic face data alone. We show that this can be extended to include the full body by building on the pipeline of Wood et al. to generate synthetic images of humans in their entirety, with ground-truth annotations for computer vision applications. In this report we describe how we construct a parametric model of the face and body, including articulated hands; our rendering pipeline to generate realistic images of humans based on this body model; an approach for training DNNs to regress a dense set of landmarks covering the entire body; and a method for fitting our body model to dense landmarks predicted from multiple views.

We propose a distributionally robust return-risk model for Markov decision processes (MDPs) under risk and reward ambiguity. The proposed model optimizes the weighted average of mean and percentile performances, and it covers the distributionally robust MDPs and the distributionally robust chance-constrained MDPs (both under reward ambiguity) as special cases. By considering that the unknown reward distribution lies in a Wasserstein ambiguity set, we derive the tractable reformulation for our model. In particular, we show that that the return-risk model can also account for risk from uncertain transition kernel when one only seeks deterministic policies, and that a distributionally robust MDP under the percentile criterion can be reformulated as its nominal counterpart at an adjusted risk level. A scalable first-order algorithm is designed to solve large-scale problems, and we demonstrate the advantages of our proposed model and algorithm through numerical experiments.

In this work we introduce reinforcement learning techniques for solving lexicographic multi-objective problems. These are problems that involve multiple reward signals, and where the goal is to learn a policy that maximises the first reward signal, and subject to this constraint also maximises the second reward signal, and so on. We present a family of both action-value and policy gradient algorithms that can be used to solve such problems, and prove that they converge to policies that are lexicographically optimal. We evaluate the scalability and performance of these algorithms empirically, demonstrating their practical applicability. As a more specific application, we show how our algorithms can be used to impose safety constraints on the behaviour of an agent, and compare their performance in this context with that of other constrained reinforcement learning algorithms.

Cooperative multi-agent reinforcement learning (c-MARL) is widely applied in safety-critical scenarios, thus the analysis of robustness for c-MARL models is profoundly important. However, robustness certification for c-MARLs has not yet been explored in the community. In this paper, we propose a novel certification method, which is the first work to leverage a scalable approach for c-MARLs to determine actions with guaranteed certified bounds. c-MARL certification poses two key challenges compared with single-agent systems: (i) the accumulated uncertainty as the number of agents increases; (ii) the potential lack of impact when changing the action of a single agent into a global team reward. These challenges prevent us from directly using existing algorithms. Hence, we employ the false discovery rate (FDR) controlling procedure considering the importance of each agent to certify per-state robustness and propose a tree-search-based algorithm to find a lower bound of the global reward under the minimal certified perturbation. As our method is general, it can also be applied in single-agent environments. We empirically show that our certification bounds are much tighter than state-of-the-art RL certification solutions. We also run experiments on two popular c-MARL algorithms: QMIX and VDN, in two different environments, with two and four agents. The experimental results show that our method produces meaningful guaranteed robustness for all models and environments. Our tool CertifyCMARL is available at https://github.com/TrustAI/CertifyCMA

Real-time individual endpoint prediction has always been a challenging task but of great clinic utility for both patients and healthcare providers. With 6,879 chronic kidney disease stage 4 (CKD4) patients as a use case, we explored the feasibility and performance of gated recurrent units with decay that models Weibull probability density function (GRU-D-Weibull) as a semi-parametric longitudinal model for real-time individual endpoint prediction. GRU-D-Weibull has a maximum C-index of 0.77 at 4.3 years of follow-up, compared to 0.68 achieved by competing models. The L1-loss of GRU-D-Weibull is ~66% of XGB(AFT), ~60% of MTLR, and ~30% of AFT model at CKD4 index date. The average absolute L1-loss of GRU-D-Weibull is around one year, with a minimum of 40% Parkes serious error after index date. GRU-D-Weibull is not calibrated and significantly underestimates true survival probability. Feature importance tests indicate blood pressure becomes increasingly important during follow-up, while eGFR and blood albumin are less important. Most continuous features have non-linear/parabola impact on predicted survival time, and the results are generally consistent with existing knowledge. GRU-D-Weibull as a semi-parametric temporal model shows advantages in built-in parameterization of missing, native support for asynchronously arrived measurement, capability of output both probability and point estimates at arbitrary time point for arbitrary prediction horizon, improved discrimination and point estimate accuracy after incorporating newly arrived data. Further research on its performance with more comprehensive input features, in-process or post-process calibration are warranted to benefit CKD4 or alike terminally-ill patients.

We propose the first joint audio-video generation framework that brings engaging watching and listening experiences simultaneously, towards high-quality realistic videos. To generate joint audio-video pairs, we propose a novel Multi-Modal Diffusion model (i.e., MM-Diffusion), with two-coupled denoising autoencoders. In contrast to existing single-modal diffusion models, MM-Diffusion consists of a sequential multi-modal U-Net for a joint denoising process by design. Two subnets for audio and video learn to gradually generate aligned audio-video pairs from Gaussian noises. To ensure semantic consistency across modalities, we propose a novel random-shift based attention block bridging over the two subnets, which enables efficient cross-modal alignment, and thus reinforces the audio-video fidelity for each other. Extensive experiments show superior results in unconditional audio-video generation, and zero-shot conditional tasks (e.g., video-to-audio). In particular, we achieve the best FVD and FAD on Landscape and AIST++ dancing datasets. Turing tests of 10k votes further demonstrate dominant preferences for our model. The code and pre-trained models can be downloaded at https://github.com/researchmm/MM-Diffusion.

IoT sensors, especially video cameras, are ubiquitously deployed around the world to perform a variety of computer vision tasks in several verticals including retail, healthcare, safety and security, transportation, manufacturing, etc. To amortize their high deployment effort and cost, it is desirable to perform multiple video analytics tasks, which we refer to as Analytical Units (AUs), off the video feed coming out of every camera. In this paper, we first show that in a multi-AU setting, changing the camera setting has disproportionate impact on different AUs performance. In particular, the optimal setting for one AU may severely degrade the performance for another AU, and further the impact on different AUs varies as the environmental condition changes. We then present Elixir, a system to enhance the video stream quality for multiple analytics on a video stream. Elixir leverages Multi-Objective Reinforcement Learning (MORL), where the RL agent caters to the objectives from different AUs and adjusts the camera setting to simultaneously enhance the performance of all AUs. To define the multiple objectives in MORL, we develop new AU-specific quality estimator values for each individual AU. We evaluate Elixir through real-world experiments on a testbed with three cameras deployed next to each other (overlooking a large enterprise parking lot) running Elixir and two baseline approaches, respectively. Elixir correctly detects 7.1% (22,068) and 5.0% (15,731) more cars, 94% (551) and 72% (478) more faces, and 670.4% (4975) and 158.6% (3507) more persons than the default-setting and time-sharing approaches, respectively. It also detects 115 license plates, far more than the time-sharing approach (7) and the default setting (0).

Variational autoencoders (VAEs) are powerful tools for learning latent representations of data used in a wide range of applications. In practice, VAEs usually require multiple training rounds to choose the amount of information the latent variable should retain. This trade-off between the reconstruction error (distortion) and the KL divergence (rate) is typically parameterized by a hyperparameter $\beta$. In this paper, we introduce Multi-Rate VAE (MR-VAE), a computationally efficient framework for learning optimal parameters corresponding to various $\beta$ in a single training run. The key idea is to explicitly formulate a response function that maps $\beta$ to the optimal parameters using hypernetworks. MR-VAEs construct a compact response hypernetwork where the pre-activations are conditionally gated based on $\beta$. We justify the proposed architecture by analyzing linear VAEs and showing that it can represent response functions exactly for linear VAEs. With the learned hypernetwork, MR-VAEs can construct the rate-distortion curve without additional training and can be deployed with significantly less hyperparameter tuning. Empirically, our approach is competitive and often exceeds the performance of multiple $\beta$-VAEs training with minimal computation and memory overheads.

Dry Eye Disease (DED) is one of the most common ocular diseases: over five percent of US adults suffer from DED. Tear film instability is a known factor for DED, and is thought to be regulated in large part by the thin lipid layer that covers and stabilizes the tear film. In order to aid eye related disease diagnosis, this work proposes a novel paradigm in using computer vision techniques to numerically analyze the tear film lipid layer (TFLL) spread. Eleven videos of the tear film lipid layer spread are collected with a micro-interferometer and a subset are annotated. A tracking algorithm relying on various pillar computer vision techniques is developed. Our method can be found at https://easytear-dev.github.io/.