Sharif Digital Repository

Server workloads operate on large volumes of data. As a result, processors executing these workloads encounter frequent L1-D misses. In a many-core processor, an L1-D miss causes a request packet to be sent to an LLC slice and a response packet to be sent back to the L1-D, which results in high overhead. While prior work targeted response packets, this work focuses on accelerating the request packets. Unlike aggressive OoO cores, simpler cores used in many-core processors cannot hide the latency of L1-D request packets. We observe that LLC slices that serve L1-D misses are strongly temporally correlated. Taking advantage of this observation, we design a simple and accurate predictor. Upon... 

We consider a basic caching system, where a single server with a database of N files (e.g. movies) is connected to a set of K users through a shared bottleneck link. Each user has a local cache memory with a size of M files. The system operates in two phases. a placement phase, where each cache memory is populated up to its size from the database, and a following delivery phase, where each user requests a file from the database, and the server is responsible for delivering the requested contents. The objective is to design the two phases to minimize the load (peak or average) of the bottleneck link. We characterize the rate-memory tradeoff of the above caching system within a factor of... 

We explore the dictionary learning problem for sparse representation when the signals are dependent. In this paper, a first-order Markovian model is considered for dependency of the signals, that has many applications especially in medical signals. It is shown that the considered dependency among the signals can degrade the performance of the existing dictionary learning algorithms. Hence, we propose a method using the Maximum Log-likelihood Estimator (MLE) and the Expectation Minimization (EM) algorithm to learn the dictionary from the signals generated under the first-order Markovian model. Simulation results show the efficiency of the proposed method in comparison with the... 

To limit the peak power consumption of a cluster, a centralized power capping system typically assigns power caps to the individual servers, which are then enforced using local capping controllers. Consequently, the performance and throughput of the servers are affected, and the runtime of jobs is extended as a result. We observe that servers in big data processing clusters often execute big data applications that have different tolerance for approximate results. To mitigate the impact of power capping, we propose a new power-Capping aware resource manager for Approximate Big data processing (CAB) that takes into consideration the minimum Quality-of-Result (QoR) of the jobs. We use... 

Estimation of 3D human pose from a single image is a challenging task because of ambiguities in projection from 3D space to the 2D image plane. A new two-stage deep convolutional neural network-based method is proposed for regressing the distance and angular difference matrices among body joints. Using the angular difference between body joints in addition to the distance between them in articulated objects such as human body can better model the structure of the shapes and increases the modelling capability of the learning method. Experimental results on HumanEva I and Human3.6M datasets show that the proposed method has substantial improvement in the mean per joint position error measure... 

Median filtering has received extensive attention from forensics analyzers, as a common content-preserving, smoothing, and denoising manipulation. We propose a detection scheme for median filtering of video sequences in compressed domain based on the singular value decomposition of the process matrix, which approximates the median filtering. Projection over some of the eigenspaces of the process matrix gives a set of features of small dimension, even as small as three, making the proposed scheme a fast and suitable detector for video median filtering. The experimental evaluations show that the proposed method outperforms the state-of-the-art detectors of median filtering, and its edge... 

The frontend stalls caused by instruction and BTB misses are a significant source of performance degradation in server processors. Prefetchers are commonly employed to mitigate frontend bottleneck. However, next-line prefetchers, which are available in server processors, are incapable of eliminating a considerable number of L1 instruction misses. Temporal instruction prefetchers, on the other hand, effectively remove most of the instruction and BTB misses but impose significant area overhead. Recently, an old idea of using BTB-directed instruction prefetching is revived to address the limitations of temporal instruction prefetchers. While this approach leads to prefetchers with low area... 

Recent research revisits pairwise-correlating data prefetching due to its extremely low overhead. Pairwise-correlating data prefetching, however, cannot accurately detect where data streams end. As a result, pairwise-correlating data prefetchers either expose low accuracy or they lose timeliness when they are performing multi-degree prefetching. In this letter, we propose a novel technique to detect where data streams end and hence, control the multi-degree prefetching in the context of pairwise-correlated prefetchers. The key idea is to have a separate metadata table that operates one step ahead of the main metadata table. This way, the runahead metadata table harnesses the degree of... 

Grammar induction is one of attractive research areas of natural language processing. Since both supervised and to some extent semi-supervised grammar induction methods require large treebanks, and for many languages, such treebanks do not currently exist, we focused our attention on unsupervised approaches. Constituent Context Model (CCM) seems to be the state of the art in unsupervised grammar induction. In this paper, we show that the performance of CCM in free word order languages (FWOLs) such as Persian is inferior to that of fixed order languages such as English. We also introduce a novel approach, called parent-based constituent context model (PCCM), and show that by using some... 

Recent evaluations of mapping systems show that lack of background knowledge, most often domain specific knowledge, is one of the key problems of mapping systems these days. In fact, at present, most state of the art systems, for the tasks of mapping large ontologies, perform not with such high values of recall (~ 30%), because they mainly rely on label and structure based similarity measures. Disregarding context knowledge in ontology mapping is another drawback that almost all current approaches suffer from. In this paper we use the semantic web as background knowledge and introduce a novel approach for capturing context knowledge from the ontology for improving mapping results. We have... 

Multiple human 3D pose estimation is a useful but challenging task in computer vison applications. The ambiguities in estimation of 2D and 3D poses of multiple persons can be verified by using multi-view frames, in which the occluded or self-occluded body parts of some persons might be visible in other camera views. But, when cameras are moving and uncalibrated, estimating the association of multiple human body parts among different camera views is a challenging task. This paper presents novel methods for multiple human 3D pose estimation and pose association in multi-view camera frames in an uncalibrated camera setup using an adversarial learning framework. The generator is a 3D pose... 

Deep Neural networks (DNNs) are getting deeper and larger which intensify the data movement and compute demands. Prior work focuses on reducing data movements and computation through exploiting sparsity and similarity. However, none of them exploit input similarity and only focus on sparsity and weight similarity. Synergistically analysing the similarity and sparsity of inputs and weights, we show that memory accesses and computations can be reduced by 5.7× and 4.1×, more than what can be decreased by exploiting only sparsity, and 3.9× and 2.1×, more than what can be decreased by exploiting only weight similarity. We propose a new data-aware compression approach, called DANA, to effectively... 

We consider the problem of tracking an unknown small target from aerial videos of medium to high altitudes. This is a challenging problem, which is even more pronounced in unavoidable scenarios of drastic camera motion and high density. To address this problem, we introduce a context-aware IoU-guided tracker (COMET) that exploits a multitask two-stream network and an offline reference proposal generation strategy. The proposed network fully exploits target-related information by multi-scale feature learning and attention modules. The proposed strategy introduces an efficient sampling strategy to generalize the network on the target and its parts without imposing extra computational... 

One of the main issues in colour image processing is changing objects' colour due to colour of illumination source. Colour constancy methods tend to modify overall image colour as if it was captured under natural light illumination. Without colour constancy, the colour would be an unreliable cue to object identity. Till now, many methods in colour constancy domain are presented. They are in two categories; statistical methods and learning-based methods. This paper presents a new statistical weighted algorithm for illuminant estimation. Weights are adjusted to highlight two key factors in the image for illuminant estimation, that is contrast and brightness. The focus was on the convex part of... 

One of the main issues in colour image processing is changing objects' colour due to colour of illumination source. Colour constancy methods tend to modify overall image colour as if it was captured under natural light illumination. Without colour constancy, the colour would be an unreliable cue to object identity. Till now, many methods in colour constancy domain are presented. They are in two categories; statistical methods and learning-based methods. This paper presents a new statistical weighted algorithm for illuminant estimation. Weights are adjusted to highlight two key factors in the image for illuminant estimation, that is contrast and brightness. The focus was on the convex part of... 

Recently, energy harvesting systems that utilize hybrid NVM-SRAM memory in their designs are introduced as a promising alternative for battery-operated systems. Since the ambient input power of an energy harvesting system fluctuates as the environmental conditions change, the system may stop the execution of programs until it receives enough energy to continue the execution. Resuming the execution of a program after the suspension may lead to data inconsistency in an energy harvesting system and threatens the correct functionality of the programs. In this article, we propose COACH, an energy-efficient consistency-aware memory scheme which guarantees the correct functionality and consistency... 

Poisson Factorization (PF) is the gold standard framework for recommendation systems with implicit feedback whose variants show state-of-the-art performance on real-world recommendation tasks. However, most of the previous work do not explicitly take into account the temporal behavior of users which is essential to recommend the right item to the right user at the right time. In this paper, we introduce a Recurrent Poisson Factorization (RPF) framework that generalizes the classical PF methods by utilizing a Poisson process for modeling the implicit feedback. RPF treats time as a natural constituent of the model, and takes important factors for recommendation into consideration to provide a... 

Recommendation plays an increasingly important role in our daily lives. Recommender systems automatically suggest items to users that might be interesting for them. Recent studies illustrate that incorporating social trust in Matrix Factorization methods demonstrably improves accuracy of rating prediction. Such approaches mainly use the trust scores explicitly expressed by users. However, it is often challenging to have users provide explicit trust scores of each other. There exist quite a few works, which propose Trust Metrics to compute and predict trust scores between users based on their interactions. In this paper, first we present how social relation can be extracted from users'... 

Recommendation plays an increasingly important role in our daily lives. Recommender systems automatically suggest items to users that might be interesting for them. Recent studies illustrate that incorporating social trust in Matrix Factorization methods demonstrably improves accuracy of rating prediction. Such approaches mainly use the trust scores explicitly expressed by users. However, it is often challenging to have users provide explicit trust scores of each other. There exist quite a few works, which propose Trust Metrics to compute and predict trust scores between users based on their interactions. In this paper, first we present how social relation can be extracted from users'... 

Poisson Factorization (PF) is the gold standard framework for recommendation systems with implicit feedback whose variants show state-of-the-art performance on real-world recommendation tasks. However, they do not explicitly take into account the temporal behavior of users which is essential to recommend the right item to the right user at the right time. In this paper, we introduce Recurrent Poisson Factorization (RPF) framework that generalizes the classical PF methods by utilizing a Poisson process for modeling the implicit feedback. RPF treats time as a natural constituent of the model, and takes important factors for recommendation into consideration to provide a rich family of...