Search

The remarkable advancements in artificial intelligence (AI), primarily driven by deep neural networks, are facing challenges surrounding unsustainable computational trajectories, limited robustness, and a lack of explainability. To develop next-generation cognitive AI systems, neuro-symbolic AI emerges as a promising paradigm, fusing neural and symbolic approaches to enhance interpretability, robustness, and trustworthiness, while facilitating learning from much less data. Recent neuro-symbolic systems have demonstrated great potential in collaborative human-AI scenarios with reasoning and cognitive capabilities. In this paper, we aim to understand the workload characteristics and potential architectures for neuro-symbolic AI. We first systematically categorize neuro-symbolic AI algorithms, and then experimentally evaluate and analyze them in terms of runtime, memory, computational operators, sparsity, and system characteristics on CPUs, GPUs, and edge SoCs. Our studies reveal that neuro-symbolic models suffer from inefficiencies on off-the-shelf hardware, due to the memory-bound nature of vector-symbolic and logical operations, complex flow control, data dependencies, sparsity variations, and limited scalability. Based on profiling insights, we suggest cross-layer optimization solutions and present a hardware acceleration case study for vector-symbolic architecture to improve the performance, efficiency, and scalability of neuro-symbolic computing. Finally, we discuss the challenges and potential future directions of neuro-symbolic AI from both system and architectural perspectives., Comment: 14 pages, 11 figures, 7 tables; IEEE Transactions on Circuits and Systems for Artificial Intelligence (TCASAI), 2024

We introduce Imagen 3, a latent diffusion model that generates high quality images from text prompts. We describe our quality and responsibility evaluations. Imagen 3 is preferred over other state-of-the-art (SOTA) models at the time of evaluation. In addition, we discuss issues around safety and representation, as well as methods we used to minimize the potential harm of our models.

Disentangling attributes of various sensory signals is central to human-like perception and reasoning and a critical task for higher-order cognitive and neuro-symbolic AI systems. An elegant approach to represent this intricate factorization is via high-dimensional holographic vectors drawing on brain-inspired vector symbolic architectures. However, holographic factorization involves iterative computation with high-dimensional matrix-vector multiplications and suffers from non-convergence problems. In this paper, we present H3DFact, a heterogeneous 3D integrated in-memory compute engine capable of efficiently factorizing high-dimensional holographic representations. H3DFact exploits the computation-in-superposition capability of holographic vectors and the intrinsic stochasticity associated with memristive-based 3D compute-in-memory. Evaluated on large-scale factorization and perceptual problems, H3DFact demonstrates superior capability in factorization accuracy and operational capacity by up to five orders of magnitude, with 5.5x compute density, 1.2x energy efficiency improvements, and 5.9x less silicon footprint compared to iso-capacity 2D designs., Comment: 2024 Design Automation and Test in Europe (DATE); The first two authors have equal contributions

With unprecedented demand for generative AI (GenAI) inference, acceleration of primitives that dominate GenAI such as general matrix-vector multiplication (GEMV) is receiving considerable attention. A challenge with GEMVs is the high memory bandwidth this primitive demands. Multiple memory vendors have proposed commercially viable processing-in-memory (PIM) prototypes that attain bandwidth boost over processor via augmenting memory banks with compute capabilities and broadcasting same command to all banks. While proposed PIM designs stand to accelerate GEMV, we observe in this work that a key impediment to truly harness PIM acceleration is deducing optimal data-placement to place the matrix in memory banks. To this end, we tease out several factors that impact data-placement and propose PIMnast methodology which, like a gymnast, balances these factors to identify data-placements that deliver GEMV acceleration. Across a spectrum of GenAI models, our proposed PIMnast methodology along with additional orchestration knobs we identify delivers up to 6.86$\times$ speedup for GEMVs (of the available 7$\times$ roofline speedup) leading to up to 5$\times$ speedup for per-token latencies.

Early detection of dental disease is crucial to prevent adverse outcomes. Today, dental X-rays are currently the most accurate gold standard for dental disease detection. Unfortunately, regular X-ray exam is still a privilege for billions of people around the world. In this paper, we ask: "Can we develop a low-cost sensing system that enables dental self-examination in the comfort of one's home?" This paper presents ToMoBrush, a dental health sensing system that explores using off-the-shelf sonic toothbrushes for dental condition detection. Our solution leverages the fact that a sonic toothbrush produces rich acoustic signals when in contact with teeth, which contain important information about each tooth's status. ToMoBrush extracts tooth resonance signatures from the acoustic signals to characterize varied dental health conditions of the teeth. We evaluate ToMoBrush on 19 participants and dental-standard models for detecting common dental problems including caries, calculus, and food impaction, achieving a detection ROC-AUC of 0.90, 0.83, and 0.88 respectively. Interviews with dental experts validate ToMoBrush's potential in enhancing at-home dental healthcare.

In this paper, we introduce a new method for querying triadic concepts through partial or complete matching of triples using an inverted index, to retrieve already computed triadic concepts that contain a set of terms in their extent, intent, and/or modus. As opposed to the approximation approach described in Ananias, this method (i) does not need to keep the initial triadic context or its three dyadic counterparts, (ii) avoids the application of derivation operators on the triple components through context exploration, and (iii) eliminates the requirement for a factorization phase to get triadic concepts as the answer to one-dimensional queries. Additionally, our solution introduces a novel metric for ranking the retrieved triadic concepts based on their similarity to a given query. Lastly, an empirical study is primarily done to illustrate the effectiveness and scalability of our approach against the approximation one. Our solution not only showcases superior efficiency, but also highlights a better scalability, making it suitable for big data scenarios.

In knowledge discovery applications, the pattern set generated from data can be tremendously large and hard to explore by analysts. In the Formal Concept Analysis (FCA) framework, there have been studies to identify important formal concepts through the stability index and other quality measures. In this paper, we introduce the Base-Equivalent Conceptual Relevance (BECR) score, a novel conceptual relevance interestingness measure for improving the identification of actionable concepts. From a conceptual perspective, the base and equivalent attributes are considered meaningful information and are highly essential to maintain the conceptual structure of concepts. Thus, the basic idea of BECR is that the more base and equivalent attributes and minimal generators a concept intent has, the more relevant it is. As such, BECR quantifies these attributes and minimal generators per concept intent. Our preliminary experiments on synthetic and real-world datasets show the efficiency of BECR compared to the well-known stability index.

Data format innovations have been critical for machine learning (ML) scaling, which in turn fuels ground-breaking ML capabilities. However, even in the presence of low-precision formats, model weights are often stored in both high-precision and low-precision during training. Furthermore, with emerging directional data formats (e.g., MX9, MX6, etc.) multiple low-precision weight copies can be required. To lower memory capacity needs of weights, we explore just-in-time quantization (JIT-Q) where we only store high-precision weights in memory and generate low-precision weights only when needed. To perform JIT-Q efficiently, in this work, we evaluate emerging processing-in-memory (PIM) technology to execute quantization. With PIM, we can offload quantization to in-memory compute units enabling quantization to be performed without incurring costly data movement while allowing quantization to be concurrent with accelerator computation. Our proposed PIM-offloaded quantization keeps up with GPU compute and delivers considerable capacity savings (up to 24\%) at marginal throughput loss (up to 2.4\%). Said memory capacity savings can unlock several benefits such as fitting larger model in the same system, reducing model parallelism requirement, and improving overall ML training efficiency.

Integrating nanophotonics with electronics promises revolutionary applications, from LiDAR to holographic displays. Although silicon photonics is maturing, realizing active nanophotonics in the ubiquitous bulk CMOS processes remains challenging. We introduce a fabless approach to embed active nanophotonics in bulk CMOS by co-designing the back-end-of-line metal layers for optical functionality. Using a 65nm CMOS process, we create plasmonic liquid crystal modulators with switching speeds 100x faster than commercial technologies. This zero-change nanophotonics method could equip mass-produced chips with optical communications, sensing and imaging. Embedding nanophotonics in the dominant electronics platform democratizes nanofabrication, spawning technologies from chip-scale LiDAR to holographic light-field displays.

Continual demand for memory bandwidth has made it worthwhile for memory vendors to reassess processing in memory (PIM), which enables higher bandwidth by placing compute units in/near-memory. As such, memory vendors have recently proposed commercially viable PIM designs. However, these proposals are largely driven by the needs of (a narrow set of) machine learning (ML) primitives. While such proposals are reasonable given the the growing importance of ML, as memory is a pervasive component, %in this work, we make there is a case for a more inclusive PIM design that can accelerate primitives across domains. In this work, we ascertain the capabilities of commercial PIM proposals to accelerate various primitives across domains. We first begin with outlining a set of characteristics, termed PIM-amenability-test, which aid in assessing if a given primitive is likely to be accelerated by PIM. Next, we apply this test to primitives under study to ascertain efficient data-placement and orchestration to map the primitives to underlying PIM architecture. We observe here that, even though primitives under study are largely PIM-amenable, existing commercial PIM proposals do not realize their performance potential for these primitives. To address this, we identify bottlenecks that arise in PIM execution and propose hardware and software optimizations which stand to broaden the acceleration reach of commercial PIM designs (improving average PIM speedups from 1.12x to 2.49x relative to a GPU baseline). Overall, while we believe emerging commercial PIM proposals add a necessary and complementary design point in the application acceleration space, hardware-software co-design is necessary to deliver their benefits broadly.

Color centers in diamonds have emerged as a leading solid-state platform for advancing quantum technologies, satisfying the DiVincenzo criteria and recently achieving a quantum advantage in secret key distribution. Recent theoretical works estimate that general-purpose quantum computing using local quantum communication networks will require millions of physical qubits to encode thousands of logical qubits, which presents a substantial challenge to the hardware architecture at this scale. To address the unanswered scaling problem, in this work, we first introduce a scalable hardware modular architecture "Quantum System-on-Chip" (QSoC) that features compact two-dimensional arrays "quantum microchiplets" (QMCs) containing tin-vacancy (SnV-) spin qubits integrated on a cryogenic application-specific integrated circuit (ASIC). We demonstrate crucial architectural subcomponents, including (1) QSoC fabrication via a lock-and-release method for large-scale heterogeneous integration; (2) a high-throughput calibration of the QSoC for spin qubit spectral inhomogenous registration; (3) spin qubit spectral tuning functionality for inhomogenous compensation; (4) efficient spin-state preparation and measurement for improved spin and optical properties. QSoC architecture supports full connectivity for quantum memory arrays in a set of different resonant frequencies and offers the possibility for further scaling the number of solid-state physical qubits via larger and denser QMC arrays and optical frequency multiplexing networking., Comment: 26 pages, 15 figures. Comments welcome

Version incompatibility issues are rampant when reusing or reproducing deep learning models and applications. Existing techniques are limited to library dependency specifications declared in PyPI. Therefore, these techniques cannot detect version issues due to undocumented version constraints or issues involving hardware drivers or OS. To address this challenge, we propose to leverage the abundant discussions of DL version issues from Stack Overflow to facilitate version incompatibility detection. We reformulate the problem of knowledge extraction as a Question-Answering (QA) problem and use a pre-trained QA model to extract version compatibility knowledge from online discussions. The extracted knowledge is further consolidated into a weighted knowledge graph to detect potential version incompatibilities when reusing a DL project. Our evaluation results show that (1) our approach can accurately extract version knowledge with 84% accuracy, and (2) our approach can accurately identify 65% of known version issues in 10 popular DL projects with a high precision (92%), while two state-of-the-art approaches can only detect 29% and 6% of these issues with 33% and 17% precision respectively., Comment: 12 pages, FSE 2023

Through our respiratory system, many viruses and diseases frequently spread and pass from one person to another. Covid-19 served as an example of how crucial it is to track down and cut back on contacts to stop its spread. There is a clear gap in finding automatic methods that can detect hand-to-face contact in complex urban scenes or indoors. In this paper, we introduce a computer vision framework, called FaceTouch, based on deep learning. It comprises deep sub-models to detect humans and analyse their actions. FaceTouch seeks to detect hand-to-face touches in the wild, such as through video chats, bus footage, or CCTV feeds. Despite partial occlusion of faces, the introduced system learns to detect face touches from the RGB representation of a given scene by utilising the representation of the body gestures such as arm movement. This has been demonstrated to be useful in complex urban scenarios beyond simply identifying hand movement and its closeness to faces. Relying on Supervised Contrastive Learning, the introduced model is trained on our collected dataset, given the absence of other benchmark datasets. The framework shows a strong validation in unseen datasets which opens the door for potential deployment., Comment: Set to be published in the PLoS ONE Journal

This paper evaluates the efficacy of recent commercial processing-in-memory (PIM) solutions to accelerate fast Fourier transform (FFT), an important primitive across several domains. Specifically, we observe that efficient implementations of FFT on modern GPUs are memory bandwidth bound. As such, the memory bandwidth boost availed by commercial PIM solutions makes a case for PIM to accelerate FFT. To this end, we first deduce a mapping of FFT computation to a strawman PIM architecture representative of recent commercial designs. We observe that even with careful data mapping, PIM is not effective in accelerating FFT. To address this, we make a case for collaborative acceleration of FFT with PIM and GPU. Further, we propose software and hardware innovations which lower PIM operations necessary for a given FFT. Overall, our optimized PIM FFT mapping, termed Pimacolaba, delivers performance and data movement savings of up to 1.38$\times$ and 2.76$\times$, respectively, over a range of FFT sizes.