arXiv subjects

Abhishek Gupta

Publications and source records attributed to Abhishek Gupta.

At least 163 records · Page 9Linked to original sources

arXiv2022

Weighted Gaussian Process Bandits for Non-stationary Environments

In this paper, we consider the Gaussian process (GP) bandit optimization problem in a non-stationary environment. To capture external changes, the black-box function is allowed to be time-varying within a reproducing kernel Hilbert space (RKHS). To this end, we develop WGP-UCB, a novel UCB-type algorithm based on weighted Gaussian process regression. A key challenge is how to cope with infinite-dimensional feature maps. To that end, we leverage kernel approximation techniques to prove a sublinear regret bound, which is the first (frequentist) sublinear regret guarantee on weighted time-varying bandits with general nonlinear rewards. This result generalizes both non-stationary linear bandits and standard GP-UCB algorithms. Further, a novel concentration inequality is achieved for weighted Gaussian process regression with general weights. We also provide universal upper bounds and weight-dependent upper bounds for weighted maximum information gains. These results are of independent interest for applications such as news ranking and adaptive pricing, where weights can be adopted to capture the importance or quality of data. Finally, we conduct experiments to highlight the favorable gains of the proposed algorithm in many cases when compared to existing methods.

cs.LG↗

arXiv2022

Half a Dozen Real-World Applications of Evolutionary Multitasking, and More

Until recently, the potential to transfer evolved skills across distinct optimization problem instances (or tasks) was seldom explored in evolutionary computation. The concept of evolutionary multitasking (EMT) fills this gap. It unlocks a population's implicit parallelism to jointly solve a set of tasks, hence creating avenues for skills transfer between them. Despite it being early days, the idea of EMT has begun to show promise in a range of real-world applications. In the backdrop of recent advances, the contribution of this paper is twofold. First, a review of several application-oriented explorations of EMT in the literature is presented; the works are assimilated into half a dozen broad categories according to their respective application domains. Each of these six categories elaborates fundamental motivations to multitask, and contains a representative experimental study (referred from the literature). Second, a set of recipes is provided showing how problem formulations of general interest, those that cut across different disciplines, could be transformed in the new light of EMT. Our discussions emphasize the many practical use-cases of EMT, and is intended to spark future research towards crafting novel algorithms for real-world deployment.

cs.NE↗

arXiv2022

Learning in Sinusoidal Spaces with Physics-Informed Neural Networks

A physics-informed neural network (PINN) uses physics-augmented loss functions, e.g., incorporating the residual term from governing partial differential equations (PDEs), to ensure its output is consistent with fundamental physics laws. However, it turns out to be difficult to train an accurate PINN model for many problems in practice. In this paper, we present a novel perspective of the merits of learning in sinusoidal spaces with PINNs. By analyzing behavior at model initialization, we first show that a PINN of increasing expressiveness induces an initial bias around flat output functions. Notably, this initial solution can be very close to satisfying many physics PDEs, i.e., falling into a local minimum of the PINN loss that only minimizes PDE residuals, while still being far from the true solution that jointly minimizes PDE residuals and the initial and/or boundary conditions. It is difficult for gradient descent optimization to escape from such a local minimum trap, often causing the training to stall. We then prove that the sinusoidal mapping of inputs, in an architecture we label as sf-PINN, is effective to increase input gradient variability, thus avoiding being trapped in such deceptive local minimum. The level of variability can be effectively modulated to match high-frequency patterns in the problem at hand. A key facet of this paper is the comprehensive empirical study that demonstrates the efficacy of learning in sinusoidal spaces with PINNs for a wide range of forward and inverse modelling problems spanning multiple physics domains.

cs.LG↗

arXiv2022

The Golden Circle: Creating Socio-technical Alignment in Content Moderation

This paper outlines a conceptual framework titled The Golden Circle that describes the roles of actors at individual, organizational, and societal levels, and their dynamics in the content moderation ecosystem. Centering harm reduction and context moderation, it argues that the ML community must attend to multimodal content moderation solutions, align their work with their organizations' goals and values, and pay attention to the ever changing social contexts in which their sociotechnical systems are embedded. This is done by accounting for the why, how, and what of content moderation from a sociological and technical lens.

cs.SI↗

arXiv2022

Combining Varied Learners for Binary Classification using Stacked Generalization

The Machine Learning has various learning algorithms that are better in some or the other aspect when compared with each other but a common error that all algorithms will suffer from is training data with very high dimensional feature set. This usually ends up algorithms into generalization error that deplete the performance. This can be solved using an Ensemble Learning method known as Stacking commonly termed as Stacked Generalization. In this paper we perform binary classification using Stacked Generalization on high dimensional Polycystic Ovary Syndrome dataset and prove the point that model becomes generalized and metrics improve significantly. The various metrics are given in this paper that also point out a subtle transgression found with Receiver Operating Characteristic Curve that was proved to be incorrect.

cs.LG↗

arXiv2022

HAA4D: Few-Shot Human Atomic Action Recognition via 3D Spatio-Temporal Skeletal Alignment

Human actions involve complex pose variations and their 2D projections can be highly ambiguous. Thus 3D spatio-temporal or 4D (i.e., 3D+T) human skeletons, which are photometric and viewpoint invariant, are an excellent alternative to 2D+T skeletons/pixels to improve action recognition accuracy. This paper proposes a new 4D dataset HAA4D which consists of more than 3,300 RGB videos in 300 human atomic action classes. HAA4D is clean, diverse, class-balanced where each class is viewpoint-balanced with the use of 4D skeletons, in which as few as one 4D skeleton per class is sufficient for training a deep recognition model. Further, the choice of atomic actions makes annotation even easier, because each video clip lasts for only a few seconds. All training and testing 3D skeletons in HAA4D are globally aligned, using a deep alignment model to the same global space, making each skeleton face the negative z-direction. Such alignment makes matching skeletons more stable by reducing intraclass variations and thus with fewer training samples per class needed for action recognition. Given the high diversity and skeletal alignment in HAA4D, we construct the first baseline few-shot 4D human atomic action recognition network without bells and whistles, which produces comparable or higher performance than relevant state-of-the-art techniques relying on embedded space encoding without explicit skeletal alignment, using the same small number of training samples of unseen classes.

cs.CV↗

arXiv2022

State of AI Ethics Report (Volume 6, February 2022)

This report from the Montreal AI Ethics Institute (MAIEI) covers the most salient progress in research and reporting over the second half of 2021 in the field of AI ethics. Particular emphasis is placed on an "Analysis of the AI Ecosystem", "Privacy", "Bias", "Social Media and Problematic Information", "AI Design and Governance", "Laws and Regulations", "Trends", and other areas covered in the "Outside the Boxes" section. The two AI spotlights feature application pieces on "Constructing and Deconstructing Gender with AI-Generated Art" as well as "Will an Artificial Intellichef be Cooking Your Next Meal at a Michelin Star Restaurant?". Given MAIEI's mission to democratize AI, submissions from external collaborators have featured, such as pieces on the "Challenges of AI Development in Vietnam: Funding, Talent and Ethics" and using "Representation and Imagination for Preventing AI Harms". The report is a comprehensive overview of what the key issues in the field of AI ethics were in 2021, what trends are emergent, what gaps exist, and a peek into what to expect from the field of AI ethics in 2022. It is a resource for researchers and practitioners alike in the field to set their research and development agendas to make contributions to the field of AI ethics.

cs.CY↗

arXiv2022

Effects of Parametric and Non-Parametric Methods on High Dimensional Sparse Matrix Representations

The semantics are derived from textual data that provide representations for Machine Learning algorithms. These representations are interpretable form of high dimensional sparse matrix that are given as an input to the machine learning algorithms. Since learning methods are broadly classified as parametric and non-parametric learning methods, in this paper we provide the effects of these type of algorithms on the high dimensional sparse matrix representations. In order to derive the representations from the text data, we have considered TF-IDF representation with valid reason in the paper. We have formed representations of 50, 100, 500, 1000 and 5000 dimensions respectively over which we have performed classification using Linear Discriminant Analysis and Naive Bayes as parametric learning method, Decision Tree and Support Vector Machines as non-parametric learning method. We have later provided the metrics on every single dimension of the representation and effect of every single algorithm detailed in this paper.

cs.LG↗

arXiv2022

Safe Model-based Off-policy Reinforcement Learning for Eco-Driving in Connected and Automated Hybrid Electric Vehicles

Connected and Automated Hybrid Electric Vehicles have the potential to reduce fuel consumption and travel time in real-world driving conditions. The eco-driving problem seeks to design optimal speed and power usage profiles based upon look-ahead information from connectivity and advanced mapping features. Recently, Deep Reinforcement Learning (DRL) has been applied to the eco-driving problem. While the previous studies synthesize simulators and model-free DRL to reduce online computation, this work proposes a Safe Off-policy Model-Based Reinforcement Learning algorithm for the eco-driving problem. The advantages over the existing literature are three-fold. First, the combination of off-policy learning and the use of a physics-based model improves the sample efficiency. Second, the training does not require any extrinsic rewarding mechanism for constraint satisfaction. Third, the feasibility of trajectory is guaranteed by using a safe set approximated by deep generative models. The performance of the proposed method is benchmarked against a baseline controller representing human drivers, a previously designed model-free DRL strategy, and the wait-and-see optimal solution. In simulation, the proposed algorithm leads to a policy with a higher average speed and a better fuel economy compared to the model-free agent. Compared to the baseline controller, the learned strategy reduces the fuel consumption by more than 21\% while keeping the average speed comparable.

cs.LG↗

arXiv2022

Discriminant Analysis in Contrasting Dimensions for Polycystic Ovary Syndrome Prognostication

A lot of prognostication methodologies have been formulated for early detection of Polycystic Ovary Syndrome also known as PCOS using Machine Learning. PCOS is a binary classification problem. Dimensionality Reduction methods impact the performance of Machine Learning to a greater extent and using a Supervised Dimensionality Reduction method can give us a new edge to tackle this problem. In this paper we present Discriminant Analysis in different dimensions with Linear and Quadratic form for binary classification along with metrics. We were able to achieve good accuracy and less variation with Discriminant Analysis as compared to many commonly used classification algorithms with training accuracy reaching 97.37% and testing accuracy of 95.92% using Quadratic Discriminant Analysis. Paper also gives the analysis of data with visualizations for deeper understanding of problem.

cs.LG↗

arXiv2022

Learning in Markov Decision Processes under Constraints

We consider reinforcement learning (RL) in Markov Decision Processes in which an agent repeatedly interacts with an environment that is modeled by a controlled Markov process. At each time step $t$, it earns a reward, and also incurs a cost-vector consisting of $M$ costs. We design model-based RL algorithms that maximize the cumulative reward earned over a time horizon of $T$ time-steps, while simultaneously ensuring that the average values of the $M$ cost expenditures are bounded by agent-specified thresholds $c^{ub}_i,i=1,2,\ldots,M$. In order to measure the performance of a reinforcement learning algorithm that satisfies the average cost constraints, we define an $M+1$ dimensional regret vector that is composed of its reward regret, and $M$ cost regrets. The reward regret measures the sub-optimality in the cumulative reward, while the $i$-th component of the cost regret vector is the difference between its $i$-th cumulative cost expense and the expected cost expenditures $Tc^{ub}_i$. We prove that the expected value of the regret vector of UCRL-CMDP, is upper-bounded as $\tilde{O}\left(T^{2\slash 3}\right)$, where $T$ is the time horizon. We further show how to reduce the regret of a desired subset of the $M$ costs, at the expense of increasing the regrets of rewards and the remaining costs. To the best of our knowledge, ours is the only work that considers non-episodic RL under average cost constraints, and derive algorithms that can~\emph{tune the regret vector} according to the agent's requirements on its cost regrets.

cs.LG↗

arXiv2021

The Co-Designed Post-Pandemic University: A Participatory and Continual Learning Approach for the Future of Work

The pandemic has shattered the traditional enclosures of learning. The post-pandemic university (PPU) will no longer be contained within the 4 walls of a lecture theatre, and finish once students have left the premises. The use of online services has now blended home and university life, and the PPU needs to reflect this. Our proposal of a continuous learning model will take advantage of the newfound omnipresence of learning, while being dynamic enough to continually adapt to the ever-evolving virus situation. Universities restricting themselves to fixed subject themes that are then forgotten once completed, will miss out on the "fresh start" presented by the virus.

cs.CY↗

arXiv2021

Fully Autonomous Real-World Reinforcement Learning with Applications to Mobile Manipulation

We study how robots can autonomously learn skills that require a combination of navigation and grasping. While reinforcement learning in principle provides for automated robotic skill learning, in practice reinforcement learning in the real world is challenging and often requires extensive instrumentation and supervision. Our aim is to devise a robotic reinforcement learning system for learning navigation and manipulation together, in an autonomous way without human intervention, enabling continual learning under realistic assumptions. Our proposed system, ReLMM, can learn continuously on a real-world platform without any environment instrumentation, without human intervention, and without access to privileged information, such as maps, objects positions, or a global view of the environment. Our method employs a modularized policy with components for manipulation and navigation, where manipulation policy uncertainty drives exploration for the navigation controller, and the manipulation module provides rewards for navigation. We evaluate our method on a room cleanup task, where the robot must navigate to and pick up items scattered on the floor. After a grasp curriculum training phase, ReLMM can learn navigation and grasping together fully automatically, in around 40 hours of autonomous real-world training.

cs.LG↗

arXiv2021

Offline RL With Resource Constrained Online Deployment

Offline reinforcement learning is used to train policies in scenarios where real-time access to the environment is expensive or impossible. As a natural consequence of these harsh conditions, an agent may lack the resources to fully observe the online environment before taking an action. We dub this situation the resource-constrained setting. This leads to situations where the offline dataset (available for training) can contain fully processed features (using powerful language models, image models, complex sensors, etc.) which are not available when actions are actually taken online. This disconnect leads to an interesting and unexplored problem in offline RL: Is it possible to use a richly processed offline dataset to train a policy which has access to fewer features in the online environment? In this work, we introduce and formalize this novel resource-constrained problem setting. We highlight the performance gap between policies trained using the full offline dataset and policies trained using limited features. We address this performance gap with a policy transfer algorithm which first trains a teacher agent using the offline dataset where features are fully available, and then transfers this knowledge to a student agent that only uses the resource-constrained features. To better capture the challenge of this setting, we propose a data collection procedure: Resource Constrained-Datasets for RL (RC-D4RL). We evaluate our transfer algorithm on RC-D4RL and the popular D4RL benchmarks and observe consistent improvement over the baseline (TD3+BC without transfer). The code for the experiments is available at https://github.com/JayanthRR/RC-OfflineRL.

cs.LG↗

arXiv2021

Adaptive Risk Minimization: Learning to Adapt to Domain Shift

A fundamental assumption of most machine learning algorithms is that the training and test data are drawn from the same underlying distribution. However, this assumption is violated in almost all practical applications: machine learning systems are regularly tested under distribution shift, due to changing temporal correlations, atypical end users, or other factors. In this work, we consider the problem setting of domain generalization, where the training data are structured into domains and there may be multiple test time shifts, corresponding to new domains or domain distributions. Most prior methods aim to learn a single robust model or invariant feature space that performs well on all domains. In contrast, we aim to learn models that adapt at test time to domain shift using unlabeled test points. Our primary contribution is to introduce the framework of adaptive risk minimization (ARM), in which models are directly optimized for effective adaptation to shift by learning to adapt on the training domains. Compared to prior methods for robustness, invariance, and adaptation, ARM methods provide performance gains of 1-4% test accuracy on a number of image classification problems exhibiting domain shift.

cs.LG↗

arXiv2021

Nash Equilibrium Control Policy against Bus-off Attacks in CAN Networks

A bus-off attack is a denial-of-service (DoS) attack which exploits error handling in the controller area network (CAN) to induce an honest node to disconnect itself from the CAN bus. This paper develops a stochastic transmission policy as a countermeasure for the controller-transmitter pair against the bus-off attack. We model this as a non-zero-sum linear-quadratic-Gaussian game between the controller-transmitter pair and the attacker. We derive Nash equilibria of the game for two different information structures of the attacker. We show that the attacker has a dominant attack strategy under both information structures. Under the dominant attack strategy, we show that the optimal control policy is linear in the system state. We further identify a necessary and a sufficient conditions on the transmission policy to have bounded average cost. The theoretical results are complemented by a detailed case study of a bus-off attack on a vehicular adaptive cruise control model.

cs.CR↗

arXiv2021

A Dynamic Watermarking Algorithm for Finite Markov Decision Problems

Dynamic watermarking, as an active intrusion detection technique, can potentially detect replay attacks, spoofing attacks, and deception attacks in the feedback channel for control systems. In this paper, we develop a novel dynamic watermarking algorithm for finite-state finite-action Markov decision processes and present upper bounds on the mean time between false alarms, and the mean delay between the time an attack occurs and when it is detected. We further compute the sensitivity of the performance of the control system as a function of the watermark. We demonstrate the effectiveness of the proposed dynamic watermarking algorithm by detecting a spoofing attack in a sensor network system.

eess.SY↗

arXiv2021

Scaling of mean skin friction in turbulent boundary layers, and fully-developed pipe and channel flows

An asymptotic $-1/2$ power-law scaling and a semi-empirical finite-$Re$ model were recently presented by Dixit et al. (2020) for skin friction in zero-pressure-gradient (ZPG) turbulent boundary layers (TBLs). In this work, a new derivation is presented which shows that these relations (i) fundamentally represent a dynamically-consistent scaling of skin friction for nominally two-dimensional ZPG TBLs and fully-developed pipes and channels, and (ii) apply individually to each of these flows. The new theoretical arguments are based on transfer of kinetic energy from mean flow to large eddies of turbulence and depend neither on flow geometry nor outer boundary condition, both of which distinguish one type of flow from the other. Using skin friction data from the literature, it is demonstrated that the finite-$Re$ model describes, as predicted by the theory, data from individual flows remarkably well; these data cover the complete range of laboratory/simulation Reynolds numbers to date. It is, however, observed that performance of the model degrades while attempting to describe data from all flows in a universal fashion. Differences in outer boundary condition and large-scale structures amongst different types of flows appear to be responsible for this degradation. An empirical correction based on Clauser's shape factor, is proposed to absorb the outer boundary condition effects into the scaling of skin friction. This correction leads to a new universal scaling and a robust, semi-empirical, universal finite-$Re$ model for skin friction in ZPG TBLs, pipes and channels. Remarkable collapse of data from all flows in the new scaling underscores the importance of a dynamically-consistent approach towards revealing universality of skin friction in wall turbulence.

physics.flu-dyn↗