An official website of the United States government
The increasing complexity of deep learning systems has pushed conventional computing technologies to their limits. While the memristor is one of the prevailing technologies for deep learning acceleration, it is only suited for classical learning layers where only two operands, namely weights and inputs, are processed simultaneously. Meanwhile, to improve the computational efficiency of deep learning for emerging applications, a variety of non-traditional layers requiring concurrent processing of many operands are becoming popular. For example, hypernetworks improve their predictive robustness by simultaneously processing weights and inputs against the application context. Two-electrode memristor grids cannot directly map emerging layers’ higher-order multiplicative neural interactions. Addressing this unmet need, we present crossbar processing using dual-gated memtransistors based on two-dimensional semiconductor MoS 2 . Unlike the memristor, the resistance states of memtransistors can be persistently programmed and can be actively controlled by multiple gate electrodes. Thus, the discussed memtransistor crossbar enables several advanced inference architectures beyond a conventional passive crossbar. For example, we show that sneak paths can be effectively suppressed in memtransistor crossbars, whereas they limit size scalability in a passive memristor crossbar. Similarly, exploiting gate terminals to suppress crossbar weights dynamically reduces biasing power by ∼20% in memtransistor crossbars for a fully connected layer of AlexNet. On emerging layers such as hypernetworks, collocating multiple operations within the same crossbar cells reduces operating power by ∼ 15 × on the considered network cases.
more »
« less
Single-Step Extraction of Transformer Attention with Dual-Gated Memtransistor Crossbars
We discuss how a dual-gated memtransistor crossbar can accelerate the extraction of the Transformer’s attention scores. A memtransistor is a novel two-dimensional material-based device that offers non-volatile programmability and gate tunability. Leveraging these attributes, we demonstrate the extraction of quadratic-order products on a single memtransistor and the single-step extraction of attention scores without inferring intermediate query/key vectors. The query/key-free processing of memtransistor-based attention scoring results in 2.37× lower energy with less than half crossbar cells.
more »
« less
- PAR ID:
- 10538088
- Publisher / Repository:
- IEEE
- Date Published:
- Journal Name:
- IEEE Electron Device Letters
- Volume:
- 45
- ISSN:
- 0741-3106
- Page Range / eLocation ID:
- 2005
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
Abstract Artificial neural networks have demonstrated superiority over traditional computing architectures in tasks such as pattern classification and learning. However, they do not measure uncertainty in predictions, and hence they can make wrong predictions with high confidence, which can be detrimental for many mission-critical applications. In contrast, Bayesian neural networks (BNNs) naturally include such uncertainty in their model, as the weights are represented by probability distributions (e.g. Gaussian distribution). Here we introduce three-terminal memtransistors based on two-dimensional (2D) materials, which can emulate both probabilistic synapses as well as reconfigurable neurons. The cycle-to-cycle variation in the programming of the 2D memtransistor is exploited to achieve Gaussian random number generator-based synapses, whereas 2D memtransistor based integrated circuits are used to obtain neurons with hyperbolic tangent and sigmoid activation functions. Finally, memtransistor-based synapses and neurons are combined in a crossbar array architecture to realize a BNN accelerator for a data classification task.more » « less
-
Model-serving systems have become increasingly popular, especially in real-time web applications. In such systems, users send queries to the server and specify the desired performance metrics (e.g., desired accuracy, latency). The server maintains a set of models (model zoo) in the back-end and serves the queries based on the specified metrics. This paper examines the security, specifically robustness against model extraction attacks, of such systems. Existing black-box attacks assume a single model can be repeatedly selected for serving inference requests. Modern inference serving systems break this assumption. Thus, they cannot be directly applied to extract a victim model, as models are hidden behind a layer of abstraction exposed by the serving system. An attacker can no longer identify which model she is interacting with. To this end, we first propose a query-efficient fingerprinting algorithm to enable the attacker to trigger any desired model consistently. We show that by using our fingerprinting algorithm, model extraction can have fidelity and accuracy scores within 1% of the scores obtained when attacking a single, explicitly specified model, as well as up to 14.6% gain in accuracy and up to 7.7% gain in fidelity compared to the naive attack. Second, we counter the proposed attack with a noise-based defense mechanism that thwarts fingerprinting by adding noise to the specified performance metrics. The proposed defense strategy reduces the attack's accuracy and fidelity by up to 9.8% and 4.8%, respectively (on medium-sized model extraction). Third, we show that the proposed defense induces a fundamental trade-off between the level of protection and system goodput, achieving configurable and significant victim model extraction protection while maintaining acceptable goodput (>80%). We implement the proposed defense in a real system with plans to open source.more » « less
-
ReRAM crossbar array as a high-parallel fast and energy-efficient structure attracts much attention, especially on the acceleration of Deep Neural Network (DNN) inference on one specific task. However, due to the high energy consumption of weight re-programming and the ReRAM cells’ low endurance problem, adapting the crossbar array for multiple tasks has not been well explored. In this paper, we propose XMA, a novel crossbar-aware shift-based mask learning method for multiple task adaption in the ReRAM crossbar DNN accelerator for the first time. XMA leverages the popular mask-based learning algorithm’s benefit to mitigate catastrophic forgetting and learn a task-specific, crossbar column-wise, and shift-based multi-level mask, rather than the most commonly used elementwise binary mask, for each new task based on a frozen backbone model. With our crossbar-aware design innovation, the required masking operation to adapt for a new task could be implemented in an existing crossbar-based convolution engine with minimal hardware/memory overhead and, more importantly, no need for power-hungry cell re-programming, unlike prior works. The extensive experimental results show that, compared with state-of-the art multiple task adaption Piggyback method [1], XMA achieves 3.19% higher accuracy on average, while saving 96.6% memory overhead. Moreover, by eliminating cell re-programming, XMA achieves ∼4.3× higher energy efficiency than Piggyback.more » « less
-
Long document re-ranking has been a challenging problem for neural re-rankers based on deep language models like BERT. Early work breaks the documents into short passage-like chunks. These chunks are independently mapped to scalar scores or latent vectors, which are then pooled into a final relevance score. These encode-and-pool methods however inevitably introduce an information bottleneck: the low dimension representations. In this paper, we propose instead to model full query-to-document interaction, leveraging the attention operation and modular Transformer re-ranker framework. First, document chunks are encoded independently with an encoder module. An interaction module then encodes the query and performs joint attention from the query to all document chunk representations. We demonstrate that the model can use this new degree of freedom to aggregate important information from the entire document. Our experiments show that this design produces effective re-ranking on two classical IR collections Robust04 and ClueWeb09, and a large-scale supervised collection MS-MARCO document ranking. CCS CONCEPTS • Information systemsmore » « less
- Free Publicly Accessible Full Text
-
Accepted Manuscript1.0
- Journal Article:
- https://doi.org/10.1109/LED.2024.3435540
