Over the course of the last several years there was a strong interest in application of modern optimal control techniques to the field of character animation. This interest was fueled by introduction of efficient learning based algorithms for policy optimization, growth in computation power, and game engine improvements. It was shown that it is possible to generate natural looking control of a character by using two ingredients. First, the simulated agent must adhere to a motion capture dataset. And second, the character aims to track the control input from the user. The paper aims at closing the gap between the researchers and users by introducing an open source implementation of physics based character control in Unity framework that has a low entry barrier and a steep learning curve.
Neural Network (NN) classifiers can assign extreme probabilities to samples that have not appeared during training (out-of-distribution samples) resulting in erroneous and unreliable predictions. One of the causes for this unwanted behaviour lies in the use of the standard softmax operator which pushes the posterior probabilities to be either zero or unity hence failing to model uncertainty. The statistical derivation of the softmax operator relies on the assumption that the distributions of the latent variables for a given class are Gaussian with known variance. However, it is possible to use different assumptions in the same derivation and attain from other families of distributions as well. This allows derivation of novel operators with more favourable properties. Here, a novel operator is proposed that is derived using $t$-distributions which are capable of providing a better description of uncertainty. It is shown that classifiers that adopt this novel operator can be more robust to out of distribution samples, often outperforming NNs that use the standard softmax operator. These enhancements can be reached with minimal changes to the NN architecture.
The technological and scientific challenges involved in the development of autonomous vehicles (AVs) are currently of primary interest for many automobile companies and research labs. However, human-controlled vehicles are likely to remain on the roads for several decades to come and may share with AVs the traffic environments of the future. In such mixed environments, AVs should deploy human-like driving policies and negotiation skills to enable smooth traffic flow. To generate automated human-like driving policies, we introduce a model-free, deep reinforcement learning approach to imitate an experienced human driver's behavior. We study a static obstacle avoidance task on a two-lane highway road in simulation (Unity). Our control algorithm receives a stochastic feedback signal from two sources: a model-driven part, encoding simple driving rules, such as lane-keeping and speed control, and a stochastic, data-driven part, incorporating human expert knowledge from driving data. To assess the similarity between machine and human driving, we model distributions of track position and speed as Gaussian processes. We demonstrate that our approach leads to human-like driving policies.
Deep learning applications require global optimization of non-convex objective functions, which have multiple local minima. The same problem is often found in physical simulations and may be resolved by the methods of Langevin dynamics with Simulated Annealing, which is a well-established approach for minimization of many-particle potentials. This analogy provides useful insights for non-convex stochastic optimization in machine learning. Here we find that integration of the discretized Langevin equation gives a coordinate updating rule equivalent to the famous Momentum optimization algorithm. As a main result, we show that a gradual decrease of the momentum coefficient from the initial value close to unity until zero is equivalent to application of Simulated Annealing or slow cooling, in physical terms. Making use of this novel approach, we propose CoolMomentum -- a new stochastic optimization method. Applying Coolmomentum to optimization of Resnet-20 on Cifar-10 dataset and Efficientnet-B0 on Imagenet, we demonstrate that it is able to achieve high accuracies.
Autonomous driving system development is critically dependent on the ability to replay complex and diverse traffic scenarios in simulation. In such scenarios, the ability to accurately simulate the vehicle sensors such as cameras, lidar or radar is essential. However, current sensor simulators leverage gaming engines such as Unreal or Unity, requiring manual creation of environments, objects and material properties. Such approaches have limited scalability and fail to produce realistic approximations of camera, lidar, and radar data without significant additional work. In this paper, we present a simple yet effective approach to generate realistic scenario sensor data, based only on a limited amount of lidar and camera data collected by an autonomous vehicle. Our approach uses texture-mapped surfels to efficiently reconstruct the scene from an initial vehicle pass or set of passes, preserving rich information about object 3D geometry and appearance, as well as the scene conditions. We then leverage a SurfelGAN network to reconstruct realistic camera images for novel positions and orientations of the self-driving vehicle and moving objects in the scene. We demonstrate our approach on the Waymo Open Dataset and show that it can synthesize realistic camera data for simulated scenarios. We also create a novel dataset that contains cases in which two self-driving vehicles observe the same scene at the same time. We use this dataset to provide additional evaluation and demonstrate the usefulness of our SurfelGAN model.
Our effort is toward unifying GAN and DRL algorithms into a unifying AI model (AGI or general-purpose AI or artificial general intelligence which has general-purpose applications to: (A) offline learning (of stored data) like GAN in (un/semi-/fully-)SL setting such as big data analytics (mining) and visualization; (B) online learning (of real or simulated devices) like DRL in RL setting (with/out environment reward) such as (real or simulated) robotics and control; Our core proposal is adding an (generative/predictive) environment model to the actor-critic (model-free) architecture which results in a model-based actor-critic architecture with temporal-differencing (TD) error and an episodic memory. The proposed AI model is similar to (model-free) DDPG and therefore it's called model-based DDPG. To evaluate it, we compare it with (model-free) DDPG by applying them both to a variety (wide range) of independent simulated robotic and control task environments in OpenAI Gym and Unity Agents. Our initial limited experiments show that DRL and GAN in model-based actor-critic results in an incremental goal-driven intellignce required to solve each task with similar performance to (model-free) DDPG. Our future focus is to investigate the proposed AI model potential to: (A) unify DRL field inside AI by producing competitive performance compared to the best of model-based (PlaNet) and model-free (D4PG) approaches; (B) bridge the gap between AI and robotics communities by solving the important problem of reward engineering with learning the reward function by demonstration.
Style variation has been a major challenge for person re-identification, which aims to match the same pedestrians across different cameras. Existing works attempted to address this problem with camera-invariant descriptor subspace learning. However, there will be more image artifacts when the difference between the images taken by different cameras is larger. To solve this problem, we propose a UnityStyle adaption method, which can smooth the style disparities within the same camera and across different cameras. Specifically, we firstly create UnityGAN to learn the style changes between cameras, producing shape-stable style-unity images for each camera, which is called UnityStyle images. Meanwhile, we use UnityStyle images to eliminate style differences between different images, which makes a better match between query and gallery. Then, we apply the proposed method to Re-ID models, expecting to obtain more style-robust depth features for querying. We conduct extensive experiments on widely used benchmark datasets to evaluate the performance of the proposed framework, the results of which confirm the superiority of the proposed model.
We present a unified representation for actionable spatial perception: 3D Dynamic Scene Graphs. Scene graphs are directed graphs where nodes represent entities in the scene (e.g. objects, walls, rooms), and edges represent relations (e.g. inclusion, adjacency) among nodes. Dynamic scene graphs (DSGs) extend this notion to represent dynamic scenes with moving agents (e.g. humans, robots), and to include actionable information that supports planning and decision-making (e.g. spatio-temporal relations, topology at different levels of abstraction). Our second contribution is to provide the first fully automatic Spatial PerceptIon eNgine(SPIN) to build a DSG from visual-inertial data. We integrate state-of-the-art techniques for object and human detection and pose estimation, and we describe how to robustly infer object, robot, and human nodes in crowded scenes. To the best of our knowledge, this is the first paper that reconciles visual-inertial SLAM and dense human mesh tracking. Moreover, we provide algorithms to obtain hierarchical representations of indoor environments (e.g. places, structures, rooms) and their relations. Our third contribution is to demonstrate the proposed spatial perception engine in a photo-realistic Unity-based simulator, where we assess its robustness and expressiveness. Finally, we discuss the implications of our proposal on modern robotics applications. 3D Dynamic Scene Graphs can have a profound impact on planning and decision-making, human-robot interaction, long-term autonomy, and scene prediction. A video abstract is available at https://youtu.be/SWbofjhyPzI
The relative $\alpha$-entropy is the R\'enyi analog of relative entropy and arises prominently in information-theoretic problems. Recent information geometric investigations on this quantity have enabled the generalization of the Cram\'{e}r-Rao inequality, which provides a lower bound for the variance of an estimator of an escort of the underlying parametric probability distribution. However, this framework remains unexamined in the Bayesian framework. In this paper, we propose a general Riemannian metric based on relative $\alpha$-entropy to obtain a generalized Bayesian Cram\'{e}r-Rao inequality. This establishes a lower bound for the variance of an unbiased estimator for the $\alpha$-escort distribution starting from an unbiased estimator for the underlying distribution. We show that in the limiting case when the entropy order approaches unity, this framework reduces to the conventional Bayesian Cram\'{e}r-Rao inequality. Further, in the absence of priors, the same framework yields the deterministic Cram\'{e}r-Rao inequality.
The Regression Tsetlin Machine (RTM) addresses the lack of interpretability impeding state-of-the-art nonlinear regression models. It does this by using conjunctive clauses in propositional logic to capture the underlying non-linear frequent patterns in the data. These, in turn, are combined into a continuous output through summation, akin to a linear regression function, however, with non-linear components and unity weights. Although the RTM has solved non-linear regression problems with competitive accuracy, the resolution of the output is proportional to the number of clauses employed. This means that computation cost increases with resolution. To reduce this problem, we here introduce integer weighted RTM clauses. Our integer weighted clause is a compact representation of multiple clauses that capture the same sub-pattern-N repeating clauses are turned into one, with an integer weight N. This reduces computation cost N times, and increases interpretability through a sparser representation. We further introduce a novel learning scheme that allows us to simultaneously learn both the clauses and their weights, taking advantage of so-called stochastic searching on the line. We evaluate the potential of the integer weighted RTM empirically using six artificial datasets. The results show that the integer weighted RTM is able to acquire on par or better accuracy using significantly less computational resources compared to regular RTMs. We further show that integer weights yield improved accuracy over real-valued ones.
Previous works on segmentation of SEM (scanning electron microscope) blood cell image ignore the semantic segmentation approach of whole-slide blood cell segmentation. In the proposed work, we address the problem of whole-slide blood cell segmentation using the semantic segmentation approach. We design a novel convolutional encoder-decoder framework along with VGG-16 as the pixel-level feature extraction model. -e proposed framework comprises 3 main steps: First, all the original images along with manually generated ground truth masks of each blood cell type are passed through the preprocessing stage. In the preprocessing stage, pixel-level labeling, RGB to grayscale conversion of masked image and pixel fusing, and unity mask generation are performed. After that, VGG16 is loaded into the system, which acts as a pretrained pixel-level feature extraction model. In the third step, the training process is initiated on the proposed model. We have evaluated our network performance on three evaluation metrics. We obtained outstanding results with respect to classwise, as well as global and mean accuracies. Our system achieved classwise accuracies of 97.45%, 93.34%, and 85.11% for RBCs, WBCs, and platelets, respectively, while global and mean accuracies remain 97.18% and 91.96%, respectively.
This paper presented a deep reinforcement learning method named Double Deep Q-networks to design an end-to-end vision-based adaptive cruise control (ACC) system. A simulation environment of a highway scene was set up in Unity, which is a game engine that provided both physical models of vehicles and feature data for training and testing. Well-designed reward functions associated with the following distance and throttle/brake force were implemented in the reinforcement learning model for both internal combustion engine (ICE) vehicles and electric vehicles (EV) to perform adaptive cruise control. The gap statistics and total energy consumption are evaluated for different vehicle types to explore the relationship between reward functions and powertrain characteristics. Compared with the traditional radar-based ACC systems or human-in-the-loop simulation, the proposed vision-based ACC system can generate either a better gap regulated trajectory or a smoother speed trajectory depending on the preset reward function. The proposed system can be well adaptive to different speed trajectories of the preceding vehicle and operated in real-time.
Automated multi-document extractive text summarization is a widely studied research problem in the field of natural language understanding. Such extractive mechanisms compute in some form the worthiness of a sentence to be included into the summary. While the conventional approaches rely on human crafted document-independent features to generate a summary, we develop a data-driven novel summary system called HNet, which exploits the various semantic and compositional aspects latent in a sentence to capture document independent features. The network learns sentence representation in a way that, salient sentences are closer in the vector space than non-salient sentences. This semantic and compositional feature vector is then concatenated with the document-dependent features for sentence ranking. Experiments on the DUC benchmark datasets (DUC-2001, DUC-2002 and DUC-2004) indicate that our model shows significant performance gain of around 1.5-2 points in terms of ROUGE score compared with the state-of-the-art baselines.
This paper uses a graphic engine to simulate a large amount of training data with free annotations. Between synthetic and real data, there is a two-level domain gap, i.e., content level and appearance level. While the latter has been widely studied, we focus on reducing the content gap in attributes like illumination and viewpoint. To reduce the problem complexity, we choose a smaller and more controllable application, vehicle re-identification (re-ID). We introduce a large-scale synthetic dataset VehicleX. Created in Unity, it contains 1,362 vehicles of various 3D models with fully editable attributes. We propose an attribute descent approach to let VehicleX approximate the attributes in real-world datasets. Specifically, we manipulate each attribute in VehicleX, aiming to minimize the discrepancy between VehicleX and real data in terms of the Fr\'echet Inception Distance (FID). This attribute descent algorithm allows content domain adaptation (DA) orthogonal to existing appearance DA methods. We mix the optimized VehicleX data with real-world vehicle re-ID datasets, and observe consistent improvement. With the augmented datasets, we report competitive accuracy. We make the dataset, engine and our codes available at https://github.com/yorkeyao/VehicleX.
We extend the Deep Galerkin Method (DGM) introduced in Sirignano and Spiliopoulos (2018)} to solve a number of partial differential equations (PDEs) that arise in the context of optimal stochastic control and mean field games. First, we consider PDEs where the function is constrained to be positive and integrate to unity, as is the case with Fokker-Planck equations. Our approach involves reparameterizing the solution as the exponential of a neural network appropriately normalized to ensure both requirements are satisfied. This then gives rise to nonlinear a partial integro-differential equation (PIDE) where the integral appearing in the equation is handled by a novel application of importance sampling. Secondly, we tackle a number of Hamilton-Jacobi-Bellman (HJB) equations that appear in stochastic optimal control problems. The key contribution is that these equations are approached in their unsimplified primal form which includes an optimization problem as part of the equation. We extend the DGM algorithm to solve for the value function and the optimal control \simultaneously by characterizing both as deep neural networks. Training the networks is performed by taking alternating stochastic gradient descent steps for the two functions, a technique inspired by the policy improvement algorithms (PIA).
Deep convolutional neural networks are known to be unstable during training at high learning rate unless normalization techniques are employed. Normalizing weights or activations allows the use of higher learning rates, resulting in faster convergence and higher test accuracy. Batch normalization requires minibatch statistics that approximate the dataset statistics but this incurs additional compute and memory costs and causes a communication bottleneck for distributed training. Weight normalization and initialization-only schemes do not achieve comparable test accuracy. We introduce a new understanding of the cause of training instability and provide a technique that is independent of normalization and minibatch statistics. Our approach treats training instability as a spatial common mode signal which is suppressed by placing the model on a channel-wise zero-mean isocline that is maintained throughout training. Firstly, we apply channel-wise zero-mean initialization of filter kernels with overall unity kernel magnitude. At each training step we modify the gradients of spatial kernels so that their weighted channel-wise mean is subtracted in order to maintain the common mode rejection condition. This prevents the onset of mean shift. This new technique allows direct training of the test graph so that training and test models are identical. We also demonstrate that injecting random noise throughout the network during training improves generalization. This is based on the idea that, as a side effect, batch normalization performs deep data augmentation by injecting minibatch noise due to the weakness of the dataset approximation. Our technique achieves higher accuracy compared to batch normalization and for the first time shows that minibatches and normalization are unnecessary for state-of-the-art training.
If an artificial intelligence aims to maximise risk-adjusted return, then under mild conditions it is disproportionately likely to pick an unethical strategy unless the objective function allows sufficiently for this risk. Even if the proportion ${\eta}$ of available unethical strategies is small, the probability ${p_U}$ of picking an unethical strategy can become large; indeed unless returns are fat-tailed ${p_U}$ tends to unity as the strategy space becomes large. We define an Unethical Odds Ratio Upsilon (${\Upsilon}$) that allows us to calculate ${p_U}$ from ${\eta}$, and we derive a simple formula for the limit of ${\Upsilon}$ as the strategy space becomes large. We give an algorithm for estimating ${\Upsilon}$ and ${p_U}$ in finite cases and discuss how to deal with infinite strategy spaces. We show how this principle can be used to help detect unethical strategies and to estimate ${\eta}$. Finally we sketch some policy implications of this work.
As digital virtual assistants become ubiquitous, it becomes increasingly important to understand the situated behaviour of users as they interact with these assistants. To this end, we introduce SIMMC, an extension to ParlAI for multi-modal conversational data collection and system evaluation. SIMMC simulates an immersive setup, where crowd workers are able to interact with environments constructed in AI Habitat or Unity while engaging in a conversation. The assistant in SIMMC can be a crowd worker or Artificial Intelligent (AI) agent. This enables both (i) a multi-player / Wizard of Oz setting for data collection, or (ii) a single player mode for model / system evaluation. We plan to open-source a situated conversational data-set collected on this platform for the Conversational AI research community.
This paper attempts to answer a central question in unsupervised learning: what does it mean to "make sense" of a sensory sequence? In our formalization, making sense involves constructing a symbolic causal theory that both explains the sensory sequence and also satisfies a set of unity conditions. The unity conditions insist that the constituents of the causal theory -- objects, properties, and laws -- must be integrated into a coherent whole. On our account, making sense of sensory input is a type of program synthesis, but it is unsupervised program synthesis. Our second contribution is a computer implementation, the Apperception Engine, that was designed to satisfy the above requirements. Our system is able to produce interpretable human-readable causal theories from very small amounts of data, because of the strong inductive bias provided by the unity conditions. A causal theory produced by our system is able to predict future sensor readings, as well as retrodict earlier readings, and impute (fill in the blanks of) missing sensory readings, in any combination. We tested the engine in a diverse variety of domains, including cellular automata, rhythms and simple nursery tunes, multi-modal binding problems, occlusion tasks, and sequence induction intelligence tests. In each domain, we test our engine's ability to predict future sensor values, retrodict earlier sensor values, and impute missing sensory data. The engine performs well in all these domains, significantly out-performing neural net baselines. We note in particular that in the sequence induction intelligence tests, our system achieved human-level performance. This is notable because our system is not a bespoke system designed specifically to solve intelligence tests, but a general-purpose system that was designed to make sense of any sensory sequence.
The question of whether deep neural networks are good at generalising beyond their immediate training experience is of critical importance for learning-based approaches to AI. Here, we consider tests of out-of-sample generalisation that require an agent to respond to never-seen-before instructions by manipulating and positioning objects in a 3D Unity simulated room. We first describe a comparatively generic agent architecture that exhibits strong performance on these tests. We then identify three aspects of the training regime and environment that make a significant difference to its performance: (a) the number of object/word experiences in the training set; (b) the visual invariances afforded by the agent's perspective, or frame of reference; and (c) the variety of visual input inherent in the perceptual aspect of the agent's perception. Our findings indicate that the degree of generalisation that networks exhibit can depend critically on particulars of the environment in which a given task is instantiated. They further suggest that the propensity for neural networks to generalise in systematic ways may increase if, like human children, those networks have access to many frames of richly varying, multi-modal observations as they learn.