Abstract

Cognitive brain mapping is enjoying its growth with the availability of large open data sharing efforts as well as the application of modern machine learning and deep learning methods. In this article, we review some of the current practices in cognitive neuroscience predominantly focusing on functional imaging and highlight the tremendous opportunities fostered by the unprecedented scale of datasets in cognitive neuroscience. We also point out challenges and limitations to keep in mind while working with these datasets.

Abstract

Integrating scalability, interoperability, and security has become crucial with the widespread adoption of Internet of Things (IoT)-enabled smart city solutions. In this context, the oneM2M provides promising technical specifications for an interoperable and secure IoT/M2M system. This paper focuses on the potential threats and their impact on oneM2M standardbased smart city deployments. Further, configurations for baseline security of the oneM2M open-source implementation, called eclipse OM2M, are presented. Recommendations are proposed based on actual on-ground tests conducted on OM2M-based smart city deployment of IIIT Hyderabad (IIIT-H) in India. The tests cover passive eavesdropping, performing replay attacks, brute force on credentials, denial of service, and analysis of access control policies. Index Terms—Eclipse OM2M, Internet of Things (IoT) security, IoT security standardisation, oneM2M, smart city

Abstract

Blockchains have proven to provide a high level of performance in terms of security and reliability for various applications like cryptocurrencies and Internet-of-Things (IoT). Prism is a recent blockchain algorithm that achieves the physical limit on throughput and latency without compromising security. In recent days, reinforcement learning approaches are investi- gated in traditional blockchains, to improve performance. In this work, we apply Deep Reinforcement Learning (DRL) to one of the promising blockchain protocols, Prism, to optimize its performance. We propose a Deep Reinforcement Learning- based Prism Blockchain (DRLPB) scheme which dynamically optimizes the parameters of the Prism blockchain and helps in achieving a better performance. In DRLPB, we apply two widely used DRL algorithms, Dueling Deep Q Networks (DDQN) and Proximal Policy Optimization (PPO). This work presents a novel approach to applying DDQN and PPO to a blockchain protocol and comparing the performance. The DRLPB scheme adapts the Prism blockchain parameters to enhance the number of votes upto 84% more than Prism, while still preserving the security and latency performance guarantees of Prism.

Abstract

Software-as-a-Service (SaaS) product companies have brought in significant changes in how we build software from architecture and engineering process perspective. SaaS products are large, dis- tributed software systems hosted in cloud and built using collabo- rating services (or micro-services). The software releases happen in days and weeks, necessitating an agile development process. Novice engineers (those who join the company fresh from college) need to become comfortable with complex systems and proficient in agile delivery with high quality, otherwise they fall behind in productivity. The paper posits that, to be successful at these SaaS product companies, the novice engineers need good modeling and design skills. While this has been for all software development, the changes driven by SaaS products have made this need more acute. Such skills will allow them to capture their feature behaviors (in context of their understanding of the larger product) in an implementation- independent manner and any knowledge gaps can be identified and bridged by their collaborators. We propose a modeling language that is easy for them to learn and use, and which has characteristics suitable for the kind of engineering work they need to do in their early years in a SaaS product company. This modeling language is based on the notion of Transition Systems. The paper demonstrates the usage and value of this language by creating a model for a real feature. The modeling language is quite general and transcends abstraction boundaries. We also present a modeling process that should be used with this language for better results. This is a short position paper that presents an idea about a new modeling language for a specific purpose (helping novice engineers design well at SaaS product companies). Validation studies for the language and the design process is a work in progress and the results will be shared in a full paper later.

Abstract

Sparse neural networks have been proven to generate efficient and better runtimes when compared to dense neural networks. Accelera- tion in runtime is better achieved with structured sparsity. However, generating an efficient sparsity structure to maintain both runtime and accuracy is a challenging task. In this paper, we implement the RBGP4 sparsity pattern derived from the Ramanujan Bipartite Graph Product (RBGP) framework on various Computer Vision tasks and test how well it performs w.r.t accuracy and runtime. Us- ing this approach, we generate structured sparse neural networks which has multiple levels of block sparsity that generates good connectivity due to the presence of Ramanujan bipartite graphs. We benchmark our approach on Semantic Segmentation and Pose Estimation tasks on an edge device (Jetson Nano 2GB) as well as server (V100) GPUs. We compare the results obtained for RBGP4 sparsity pattern with the unstructured and block sparsity patterns. When compared to sparsity patterns like unstructured and block, we obtained significant speedups while maintaining accuracy. KEYWORDS Sparse Neural Networks, Ramanujan Graphs, Semantic Segmenta- tion, Pose Detection ACM Reference Format: Thejasvi Konduru, Furqan Ahmed Shaik, Girish Varma, and Kishore Kotha- palli. 2023. Accelerating Computer Vision Tasks on GPUs using Ramanujan Graph Product Framework. In 6th Joint International Conference on Data Science & Management of Data (10th ACM IKDD CODS and 28th COMAD) (CODS-COMAD 2023), January 4–7, 2023, Mumbai, India. ACM, New York, NY, USA, 5 pages. https://doi.org/10.1145/3570991.3571044

Abstract

Invertible convolutions have been an essential element for building expressive normalizing flow-based generative models since their introduction in Glow. Several attempts have been made to design invertible k × k convolutions that are efficient in training and sampling passes. Though these attempts have improved the expressivity and sampling efficiency, they severely lagged behind Glow which used only 1×1 convolutions in terms of sampling time. Also, many of the approaches mask a large number of parameters of the underlying convolution, resulting in lower expressivity on a fixed run-time budget. We propose a k × k convolutional layer and Deep Normalizing Flow architecture which i.) has a fast parallel inversion algorithm with running time O(nk2 ) (n is height and width of the input image and k is kernel size), ii.) masks the minimal amount of learnable parameters in a layer. iii.) gives better forward pass and sampling times comparable to other k ×k convolution-based models on real-world benchmarks. We provide an implementation of the proposed parallel algorithm for sampling using our invertible convolutions on GPUs. Benchmarks on CIFAR-10, ImageNet, and CelebA datasets show comparable performance to previous works regarding bits per dimension while significantly improving the sampling time.

Abstract

We investigate whether it is possible to teleport the coherence of an unknown quantum state from Alice to Bob by communicating lesser number of classical bits in comparison to what is required in teleporting an unknown quantum state. We find that we cannot do perfect teleportation of coherence with one bit of classical communication for an arbitrary qubit. However, we find that if the qubit is chosen from equatorial and polar circles, then teleportation of coherence will be possible with transfer of one cbit of information if we have maximally entangled states as a shared resource. In the case of resource being a non maximally entangled state, we can teleport quantum coherence with a certain probability of success. In a general teleportation protocol for coherence, we derive a compact formula for the final state at Bob’s lab in terms of composition of the completely positive maps corresponding to the shared resource state and joint POVM performed by Alice on her qubit and the unknown state. With the help of this formula, we investigate the amount of coherence teleported when the resource shared state is the maximally entangled mixed state and Werner state for the unknown pure as well as mixed states state at Alice’s lab. In particular, we show that when the Werner state becomes separable then also the amount of teleported coherence is non-zero, implying the possibility of teleportation of coherence without entanglement. We have also shown that teleportation of coherence of an arbitrary state with real matrix elements is exactly possible with the help of maximally entangled state as a resource.

Abstract

We consider the problem of learning tensors from partial observa- tions with structural constraints, under the low-rank assumption. To this end, we propose a general convex low-rank regularizer, pa- rameterized by linear maps, that extends the existing regularizers. Different choices of the linear maps lead to different convex low- rank regularizers. The resulting problem is called the generalized structured low-rank tensor learning problem. To solve this problem, we use duality theory to reformulate it into simpler sub-problems. The dual problem contains a rich geometric structure, which we exploit to develop first-order and second-order Riemannian opti- mization algorithms. The associated duality gap is derived, and it is shown to be zero. Moreover, we experimentally verify the cor- rectness of our algorithm on several special cases of the proposed general framework.

Abstract

Extreme multilabel classification or XML, in short, has emerged as a new subtopic of interest in machine learning. Compared to traditional multilabel classification, here the number of labels is extremely large, hence the name extreme multilabel classification. Using classical one versus all classification wont scale in this case due to large number of labels, same is true for any other classifiers. Embedding of labels as well as features into smaller label space is an essential first step. Moreover, other issues include existance of head and tail labels, where tail labels are labels which exist in relatively smaller number of given samples. The existence of tail labels creates issues during embedding. This area has invited application of wide range of approaches ranging from bit compression motivated from compressed sensing, tree based embeddings, deep learning based latent space embedding including using attention weights, linear algebra based embeddings such as SVD, clustering, hashing, to name a few. The community has come up with a useful set of metrics to identify the correctly the prediction for head or tail labels. Keywords: extreme classification, head and tail labels, compressed sensing, deep learning, attention

Abstract

Various studies have shown the advantages of using Machine Learning (ML) techniques for analog and digital IC design automation and optimization. Data scarcity is still an issue for electronic designs, while training highly accurate ML models. This work proposes generating and evaluating artificial data using generative adversarial networks (GANs) for circuit data to aid and improve the accuracy of ML models trained with a small training data set. The training data is obtained by various simulations in the Cadence Virtuoso, HSPICE, and Microcap design environment with TSMC 180nm and 22nm CMOS technology nodes. The artificial data is generated and tested for an appropriate set of analog and digital circuits. The experimental results show that the proposed artificial data generation significantly improves ML models and reduces the percentage error by more than 50% of the original percentage error, which were previously trained with insufficient data. Furthermore, this research aims to contribute to the extensive application of AI/ML in the field of VLSI design and technology by relieving the training data availability-related challenges.