Related papers: A Low-Footprint Class Loading Mechanism for Embedd…

Continual learning methods are known to suffer from catastrophic forgetting, a phenomenon that is particularly hard to counter for methods that do not store exemplars of previous tasks. Therefore, to reduce potential drift in the feature…

As an increasing number of software systems reach unprecedented scale, relying solely on code-level abstractions is becoming impractical. While architectural abstractions offer a means to manage these systems, maintaining their consistency…

Devirtualization is a compiler optimization that replaces indirect (virtual) function calls with direct calls. It is particularly effective in object-oriented languages, such as Java or C++, in which virtual methods are typically abundant.…

Many Java applications instantiate objects within the Java heap that are persistent but seldom if ever referenced by the application. Examples include strings, such as error messages, and collections of value objects that are preloaded for…

Although prior art has demonstrated negligible accuracy drop in sub-byte quantization -- where weights and/or activations are represented by less than 8 bits -- popular SIMD instructions of CPUs do not natively support these datatypes.…

Memory disaggregation has recently been adopted in data centers to improve resource utilization, motivated by cost and sustainability. Recent studies on large-scale HPC facilities have also highlighted memory underutilization. A promising…

Continual learning of deep neural networks is a key requirement for scaling them up to more complex applicative scenarios and for achieving real lifelong learning of these architectures. Previous approaches to the problem have considered…

Trying to cope with the constantly growing number of cores per processor, hardware architects are experimenting with modular non-cache-coherent architectures. Such architectures delegate the memory coherency to the software. On the…

Continual learning refers to the ability to acquire and transfer knowledge without catastrophically forgetting what was previously learned. In this work, we consider \emph{few-shot} continual learning in classification tasks, and we propose…

Class-incremental learning of deep networks sequentially increases the number of classes to be classified. During training, the network has only access to data of one task at a time, where each task contains several classes. In this…

Public cloud providers seek to meet stringent performance requirements and low hardware cost. A key driver of performance and cost is main memory. Memory pooling promises to improve DRAM utilization and thereby reduce costs. However,…

We describe verification techniques for embedded memory systems using efficient memory modeling (EMM), without explicitly modeling each memory bit. We extend our previously proposed approach of EMM in Bounded Model Checking (BMC) for a…

As the need for neural network-based applications to become more accurate and powerful grows, so too does their size and memory footprint. With embedded devices, whose cache and RAM are limited, this growth hinders their ability to leverage…

Virtual machines (VM) are widely used to host and isolate software modules. However, extremely small memory and low-energy budgets have so far prevented wide use of VMs on typical microcontroller-based IoT devices. In this paper, we explore…

We propose an approach to Multitask Learning (MTL) to make deep learning models faster and lighter for applications in which multiple tasks need to be solved simultaneously, which is particularly useful in embedded, real-time systems. We…

There is increasing demand to bring machine learning capabilities to low power devices. By integrating the computational power of machine learning with the deployment capabilities of low power devices, a number of new applications become…

This paper investigates the energy-aware virtual machine (VM) allocation problems in clouds along characteristics: multiple resources, fixed interval time and non-preemption of virtual machines. Many previous works have been proposed to use…

Return caching is a recent strategy that enables efficient minibatch training with multistep estimators (e.g. the {\lambda}-return) for deep reinforcement learning. By precomputing return estimates in sequential batches and then storing the…

Intermediate task transfer learning can greatly improve model performance. If, for example, one has little training data for emotion detection, first fine-tuning a language model on a sentiment classification dataset may improve performance…

We give a more space-efficient implementation of adaptive mergesort: Virtual-Memory Powersort. Using internal buffering techniques, we significantly reduce the memory consumption of the algorithm; specifically, for sorting $n$ objects the…