arXiv subjects

Muhammad Umer

Publications and source records attributed to Muhammad Umer.

47 records · Page 3Linked to original sources

arXiv2022

A norm minimization-based convex vector optimization algorithm

We propose an algorithm to generate inner and outer polyhedral approximations to the upper image of a bounded convex vector optimization problem. It is an outer approximation algorithm and is based on solving norm-minimizing scalarizations. Unlike Pascolleti-Serafini scalarization used in the literature for similar purposes, it does not involve a direction parameter. Therefore, the algorithm is free of direction-biasedness. We also propose a modification of the algorithm by introducing a suitable compact subset of the upper image, which helps in proving for the first time the finiteness of an algorithm for convex vector optimization. The computational performance of the algorithms is illustrated using some of the benchmark test problems, which shows promising results in comparison to a similar algorithm that is based on Pascoletti-Serafini scalarization.

math.OC↗

arXiv2022

False Memory Formation in Continual Learners Through Imperceptible Backdoor Trigger

In this brief, we show that sequentially learning new information presented to a continual (incremental) learning model introduces new security risks: an intelligent adversary can introduce small amount of misinformation to the model during training to cause deliberate forgetting of a specific task or class at test time, thus creating "false memory" about that task. We demonstrate such an adversary's ability to assume control of the model by injecting "backdoor" attack samples to commonly used generative replay and regularization based continual learning approaches using continual learning benchmark variants of MNIST, as well as the more challenging SVHN and CIFAR 10 datasets. Perhaps most damaging, we show this vulnerability to be very acute and exceptionally effective: the backdoor pattern in our attack model can be imperceptible to human eye, can be provided at any point in time, can be added into the training data of even a single possibly unrelated task and can be achieved with as few as just 1\% of total training dataset of a single task.

cs.LG↗

arXiv2021

Floquet higher-order Weyl and nexus semimetals

This work reports the general design and characterization of two exotic, anomalous nonequilibrium topological phases. In equilibrium systems, the Weyl nodes or the crossing points of nodal lines may become the transition points between higher-order and first-order topological phases defined on two-dimensional slices, thus featuring both hinge Fermi arc and surface Fermi arc. We advance this concept by presenting a strategy to obtain, using time-sequenced normal insulator phases only, Floquet higher-order Weyl semimetals and Floquet higher-order nexus semimetals, where the concerned topological singularities in the three-dimensional Brillouin zone border anomalous two-dimensional higher-order Floquet phases. The fascinating topological phases we obtain are previously unknown and can be experimentally studied using, for example, a three-dimensional lattice of coupled ring resonators.

cond-mat.mes-hall↗

arXiv2021

Dynamical characterization of Weyl nodes in Floquet Weyl semimetal phases

Due to studies in nonequilibrium (periodically-driven) topological matter, it is now understood that some topological invariants used to classify equilibrium states of matter do not suffice to describe their nonequilibrium counterparts. Indeed, in Floquet systems the additional gap arising from the periodicity of the quasienergy Brillouin zone often leads to unique topological phenomena without equilibrium analogues. In the context of Floquet Weyl semimetal, Weyl points may be induced at both quasienergy zero and $π/T$ ($T$ being the driving period) and these two types of Weyl points can be very close to each other in the momentum space. Because of their momentum-space proximity, the chirality of each individual Weyl point may become hard to characterize in both theory and experiments, thus making it challenging to determine the system's overall topology. In this work, inspired by the construction of dynamical winding numbers in Floquet Chern insulators, we propose a dynamical invariant capable of characterizing and distinguishing between Weyl points at different quasienergy values, thus advancing one step further in the topological characterization of Floquet Weyl semimetals. To demonstrate the usefulness of such a dynamical topological invariant, we consider a variant of the periodically kicked Harper model (the very first model in studies of Floquet topological phases) that exhibits many Weyl points, with the number of Weyl points rising unlimitedly with the strength of some system parameters. Furthermore, we investigate the two-terminal transport signature associated with the Weyl points. Theoretical findings of this work pave the way for experimentally probing the rich topological band structures of some seemingly simple Floquet semimetal systems.

cond-mat.mes-hall↗

arXiv2021

Adversarial Targeted Forgetting in Regularization and Generative Based Continual Learning Models

Continual (or "incremental") learning approaches are employed when additional knowledge or tasks need to be learned from subsequent batches or from streaming data. However these approaches are typically adversary agnostic, i.e., they do not consider the possibility of a malicious attack. In our prior work, we explored the vulnerabilities of Elastic Weight Consolidation (EWC) to the perceptible misinformation. We now explore the vulnerabilities of other regularization-based as well as generative replay-based continual learning algorithms, and also extend the attack to imperceptible misinformation. We show that an intelligent adversary can take advantage of a continual learning algorithm's capabilities of retaining existing knowledge over time, and force it to learn and retain deliberately introduced misinformation. To demonstrate this vulnerability, we inject backdoor attack samples into the training data. These attack samples constitute the misinformation, allowing the attacker to capture control of the model at test time. We evaluate the extent of this vulnerability on both rotated and split benchmark variants of the MNIST dataset under two important domain and class incremental learning scenarios. We show that the adversary can create a "false memory" about any task by inserting carefully-designed backdoor samples to the test instances of that task thereby controlling the amount of forgetting of any task of its choosing. Perhaps most importantly, we show this vulnerability to be very acute and damaging: the model memory can be easily compromised with the addition of backdoor samples into as little as 1\% of the training data, even when the misinformation is imperceptible to human eye.

cs.LG↗

arXiv2020

Comparative Analysis of Extreme Verification Latency Learning Algorithms

One of the more challenging real-world problems in computational intelligence is to learn from non-stationary streaming data, also known as concept drift. Perhaps even a more challenging version of this scenario is when -- following a small set of initial labeled data -- the data stream consists of unlabeled data only. Such a scenario is typically referred to as learning in initially labeled nonstationary environment, or simply as extreme verification latency (EVL). Because of the very challenging nature of the problem, very few algorithms have been proposed in the literature up to date. This work is a very first effort to provide a review of some of the existing algorithms (important/prominent) in this field to the research community. More specifically, this paper is a comprehensive survey and comparative analysis of some of the EVL algorithms to point out the weaknesses and strengths of different approaches from three different perspectives: classification accuracy, computational complexity and parameter sensitivity using several synthetic and real world datasets.

cs.LG↗

arXiv2020

Comparison Analysis of Tree Based and Ensembled Regression Algorithms for Traffic Accident Severity Prediction

Rapid increase of traffic volume on urban roads over time has changed the traffic scenario globally. It has also increased the ratio of road accidents that can be severe and fatal in the worst case. To improve traffic safety and its management on urban roads, there is a need for prediction of severity level of accidents. Various machine learning models are being used for accident prediction. In this study, tree based ensemble models (Random Forest, AdaBoost, Extra Tree, and Gradient Boosting) and ensemble of two statistical models (Logistic Regression Stochastic Gradient Descent) as voting classifiers are compared for prediction of road accident severity. Significant features that are strongly correlated with the accident severity are identified by Random Forest. Analysis proved Random Forest as the best performing model with highest classification results with 0.974 accuracy, 0.954 precision, 0.930 recall and 0.942 F-score using 20 most significant features as compared to other techniques classification of road accidents severity.

cs.OH↗

arXiv2020

Counter-propagating edge states in Floquet topological insulating phases

Nonequilibrium Floquet topological phases due to periodic driving are known to exhibit rich and interesting features with no static analogs. Various known topological invariants usually proposed to characterize static topological systems often fail to fully characterize Floquet topological phases. This fact has motivated extensive studies of Floquet topological phases to better understand nonequilibrium topological phases and to explore their possible applications. Here we present a theoretically simple Floquet topological insulating system that may possess an arbitrary number of counter-propagating chiral edge states. Further investigation into our system reveals another related feature by tuning the same set of system parameters, namely, the emergence of almost flat (dispersionless) edge modes. In particular, we employ two-terminal conductance and dynamical winding numbers to characterize counter-propagating chiral edge states. We further demonstrate the robustness of such edge states against symmetry preserving disorder. Finally, we identify an emergent chiral symmetry at certain sub-regimes of the Brillouin zone that can explain the presence of almost flat edge modes. Our results have exposed more interesting possibilities in Floquet topological matter.

cond-mat.mes-hall↗

arXiv2020

Targeted Forgetting and False Memory Formation in Continual Learners through Adversarial Backdoor Attacks

Artificial neural networks are well-known to be susceptible to catastrophic forgetting when continually learning from sequences of tasks. Various continual (or "incremental") learning approaches have been proposed to avoid catastrophic forgetting, but they are typically adversary agnostic, i.e., they do not consider the possibility of a malicious attack. In this effort, we explore the vulnerability of Elastic Weight Consolidation (EWC), a popular continual learning algorithm for avoiding catastrophic forgetting. We show that an intelligent adversary can bypass the EWC's defenses, and instead cause gradual and deliberate forgetting by introducing small amounts of misinformation to the model during training. We demonstrate such an adversary's ability to assume control of the model via injection of "backdoor" attack samples on both permuted and split benchmark variants of the MNIST dataset. Importantly, once the model has learned the adversarial misinformation, the adversary can then control the amount of forgetting of any task. Equivalently, the malicious actor can create a "false memory" about any task by inserting carefully-designed backdoor samples to any fraction of the test instances of that task. Perhaps most damaging, we show this vulnerability to be very acute; neural network memory can be easily compromised with the addition of backdoor samples into as little as 1% of the training data of even a single task.

cs.LG↗

arXiv2018

Direct prediction of corner state configurations from edge winding numbers in 2D and 3D chiral-symmetric lattice systems

Higher-order topological phases feature topologically protected boundary states in lower dimensions. Specifically, the zero-dimensional corner states are protected by the $d$th-order topology of a $d$-dimension system. In this work, we propose to predict different configurations of corner states from winding numbers defined for one-dimensional edges of the system. We first demonstrate the winding number characterization with a generalized two-dimensional square lattice belonging to the BDI symmetry class. In addition to the second-order topological insulating phase, the system may also be a nodal point semimetal or a weak topological insulator with topologically protected one-dimensional edge states coexisting with the corner states at zero energy. A three-dimensional cubic lattice with richer configurations of corner states is also studied. We further discuss several experimental implementations of our models with photonic lattices or electric circuits.

cond-mat.mes-hall↗

arXiv2005

A Semantic Grid-based E-Learning Framework (SELF)

E-learning can be loosely defined as a wide set of applications and processes, which uses available electronic media (and tools) to deliver vocational education and training. With its increasing recognition as an ubiquitous mode of instruction and interaction in the academic as well as corporate world, the need for a scaleable and realistic model is becoming important. In this paper we introduce SELF; a Semantic grid-based E-Learning Framework. SELF aims to identify the key-enablers in a practical grid-based E-learning environment and to minimize technological reworking by proposing a well-defined interaction plan among currently available tools and technologies. We define a dichotomy with E-learning specific application layers on top and semantic grid-based support layers underneath. We also map the latest open and freeware technologies with various components in SELF.

cs.DC↗