Mazharul Islam

Mazharul Islam, Sunpreet S. Arora, Rahul Chatterjee et al.

Secure multi-party computation (MPC) techniques can be used to provide data privacy when users query deep neural network (DNN) models hosted on a public cloud. State-of-the-art MPC techniques can be directly leveraged for DNN models that use simple activation functions such as ReLU. However, these techniques are ineffective and/or inefficient for the complex and highly non-linear activation functions used in cutting-edge DNN models. We present Compact, which produces piece-wise polynomial approximations of complex AFs to enable their efficient use with state-of-the-art MPC techniques. Compact neither requires nor imposes any restriction on model training and results in near-identical model accuracy. To achieve this, we design Compact with input density awareness and use an application-specific simulated annealing type optimization to generate computationally more efficient approximations of complex AFs. We extensively evaluate Compact on four different machine-learning tasks with DNN architectures that use popular complex AFs silu, gelu, and mish. Our experimental results show that Compact incurs negligible accuracy loss while being 2x-5x computationally more efficient than state-of-the-art approaches for DNN models with large number of hidden layers. Our work accelerates easy adoption of MPC techniques to provide user data privacy even when the queried DNN models consist of a number of hidden layers and trained over complex AFs.

Chenglong Song, Mazharul Islam, Lin Wang et al.

Conditional density estimation (CDE) is a fundamental task in machine learning that aims to model the full conditional law $\mathbb{P}(\mathbf{y} \mid \mathbf{x})$, beyond mere point prediction (e.g., mean, mode). A core challenge is free-form density estimation, capturing distributions that exhibit multimodality, asymmetry, or topological complexity without restrictive assumptions. However, prevailing methods typically estimate the probability density function (PDF) directly, which is mathematically ill-posed: differentiating the empirical distribution amplifies random fluctuations inherent in finite datasets, necessitating strong inductive biases that limit expressivity and fail when violated. We propose a CDF-first framework that circumvents this issue by estimating the cumulative distribution function (CDF), a stable and well-posed target, and then recovering the PDF via differentiation of the learned smooth CDF. Parameterizing the CDF with a Smooth Min-Max (SMM) network, our framework guarantees valid PDFs by construction, enables tractable approximate likelihood training, and preserves complex distributional shapes. For multivariate outputs, we use an autoregressive decomposition with SMM factors. Experiments demonstrate our approach outperforms state-of-the-art density estimators on a range of univariate and multivariate tasks.

Bing Chen, Mazharul Islam, Jisuo Gao et al.

Conditional density estimation (CDE) is the task of estimating the probability of an event conditioned on some inputs. A neural network (NN) can also be used to compute the output distribution for continuous-domain, which can be viewed as an extension of regression task. Nevertheless, it is difficult to explicitly approximate a distribution without knowing the information of its general form a priori. In order to fit an arbitrary conditional distribution, discretizing the continuous domain into bins is an effective strategy, as long as we have sufficiently narrow bins and very large data. However, collecting enough data is often hard to reach and falls far short of that ideal in many circumstances, especially in multivariate CDE for the curse of dimensionality. In this paper, we demonstrate the benefits of modeling free-form conditional distributions using a deconvolution-based neural net framework, coping with data deficiency problems in discretization. It has the advantage of being flexible but also takes advantage of the hierarchical smoothness offered by the deconvolution layers. We compare our method to a number of other density-estimation approaches and show that our Deconvolutional Density Network (DDN) outperforms the competing methods on many univariate and multivariate tasks. The code of DDN is available at https://github.com/NBICLAB/DDN.

Bijeeta Pal, Mazharul Islam, Marina Sanusi et al.

Credential stuffing attacks use stolen passwords to log into victim accounts. To defend against these attacks, recently deployed compromised credential checking (C3) services provide APIs that help users and companies check whether a username, password pair is exposed. These services however only check if the exact password is leaked, and therefore do not mitigate credential tweaking attacks - attempts to compromise a user account with variants of a user's leaked passwords. Recent work has shown credential tweaking attacks can compromise accounts quite effectively even when the credential stuffing countermeasures are in place. We initiate work on C3 services that protect users from credential tweaking attacks. The core underlying challenge is how to identify passwords that are similar to their leaked passwords while preserving honest clients' privacy and also preventing malicious clients from extracting breach data from the service. We formalize the problem and explore ways to measure password similarity that balance efficacy, performance, and security. Based on this study, we design "Might I Get Pwned" (MIGP), a new kind of breach alerting service. Our simulations show that MIGP reduces the efficacy of state-of-the-art 1000-guess credential tweaking attacks by 94%. MIGP preserves user privacy and limits potential exposure of sensitive breach entries. We show that the protocol is fast, with response time close to existing C3 services. We worked with Cloudflare to deploy MIGP in practice.

Deokgun Park, Md Ashaduzzaman Rubel Mondol, Aishwarya Pothula et al.

Despite recent advances of AI research in many application-specific domains, we do not know how to build a human-level artificial intelligence (HLAI). We conjecture that learning from others' experience with the language is the essential characteristic that distinguishes human intelligence from the rest. Humans can update the action-value function with the verbal description as if they experience states, actions, and corresponding rewards sequences firsthand. In this paper, we present a classification of intelligence according to how individual agents learn and propose a definition and a test for HLAI. The main idea is that language acquisition without explicit rewards can be a sufficient test for HLAI.

Mazharul Islam, Sazzadur Rahaman, Na Meng et al.

Spring security is tremendously popular among practitioners for its ease of use to secure enterprise applications. In this paper, we study the application framework misconfiguration vulnerabilities in the light of Spring security, which is relatively understudied in the existing literature. Towards that goal, we identify 6 types of security anti-patterns and 4 insecure vulnerable defaults by conducting a measurement-based approach on 28 Spring applications. Our analysis shows that security risks associated with the identified security anti-patterns and insecure defaults can leave the enterprise application vulnerable to a wide range of high-risk attacks. To prevent these high-risk attacks, we also provide recommendations for practitioners. Consequently, our study has contributed one update to the official Spring security documentation while other security issues identified in this study are being considered for future major releases by Spring security community.

Aftab Anjum, Mazharul Islam, Lin Wang

In sentiment classification, the enormous amount of textual data, its immense dimensionality, and inherent noise make it extremely difficult for machine learning classifiers to extract high-level and complex abstractions. In order to make the data less sparse and more statistically significant, the dimensionality reduction techniques are needed. But in the existing dimensionality reduction techniques, the number of components needs to be set manually which results in loss of the most prominent features, thus reducing the performance of the classifiers. Our prior work, i.e., Term Presence Count (TPC) and Term Presence Ratio (TPR) have proven to be effective techniques as they reject the less separable features. However, the most prominent and separable features might still get removed from the initial feature set despite having higher distributions among positive and negative tagged documents. To overcome this problem, we have proposed a new framework that consists of two-dimensionality reduction techniques i.e., Sentiment Term Presence Count (SentiTPC) and Sentiment Term Presence Ratio (SentiTPR). These techniques reject the features by considering term presence difference for SentiTPC and ratio of the distribution distinction for SentiTPR. Additionally, these methods also analyze the total distribution information. Extensive experimental results exhibit that the proposed framework reduces the feature dimension by a large scale, and thus significantly improve the classification performance.

Mazharul Islam, Shuangrong Liu, Lin Wang et al.

Floating centroid method (FCM) offers an efficient way to solve a fixed-centroid problem for the neural network classifiers. However, evolutionary computation as its optimization method restrains the FCM to achieve satisfactory performance for different neural network structures, because of the high computational complexity and inefficiency. Traditional gradient-based methods have been extensively adopted to optimize the neural network classifiers. In this study, a gradient-based floating centroid (GDFC) method is introduced to address the fixed centroid problem for the neural network classifiers optimized by gradient-based methods. Furthermore, a new loss function for optimizing GDFC is introduced. The experimental results display that GDFC obtains promising classification performance than the comparison methods on the benchmark datasets.

Mazharul Islam, MD. Nazmuddoha Ansary, Novia Nurain et al.

The recent emergence of new vulnerabilities is an epoch-making problem in the complex world of website security. Most of the websites are failing to keep updating to tackle their websites from these new vulnerabilities leaving without realizing the weakness of the websites. As a result, when cyber-criminals scour such vulnerable old version websites, the scanner will represent a set of vulnerabilities. Once found, these vulnerabilities are then exploited to steal data, distribute malicious content, or inject defacement and spam content into the vulnerable websites. Furthermore, a combination of different vulnerabilities is able to cause more damages than anticipation. Therefore, in this paper, we endeavor to find connections among various vulnerabilities such as cross-site scripting, local file inclusion, remote file inclusion, buffer overflow CSRF, etc. To do so, we develop a Finite State Machine (FSM) attacking model, which analyzes a set of vulnerabilities towards the road to finding connections. We demonstrate the efficacy of our model by applying it to the set of vulnerabilities found on two live websites.