scispace - formally typeset
Search or ask a question
Topic

Encoding (memory)

About: Encoding (memory) is a research topic. Over the lifetime, 7547 publications have been published within this topic receiving 120214 citations. The topic is also known as: memory encoding & encoding of memories.


Papers
More filters
Proceedings ArticleDOI
11 Jun 2021
TL;DR: HR-NAS as mentioned in this paper adopts a multi-branch architecture that provides convolutional encoding of multiple feature resolutions and proposes an efficient fine-grained search strategy to train HR-NAS, which effectively explores the search space, and finds optimal architectures given various tasks and computation resources.
Abstract: High-resolution representations (HR) are essential for dense prediction tasks such as segmentation, detection, and pose estimation. Learning HR representations is typically ignored in previous Neural Architecture Search (NAS) methods that focus on image classification. This work proposes a novel NAS method, called HR-NAS, which is able to find efficient and accurate networks for different tasks, by effectively encoding multiscale contextual information while maintaining high-resolution representations. In HR-NAS, we renovate the NAS search space as well as its searching strategy. To better encode multiscale image contexts in the search space of HR-NAS, we first carefully design a lightweight transformer, whose computational complexity can be dynamically changed with respect to different objective functions and computation budgets. To maintain high-resolution representations of the learned networks, HR-NAS adopts a multi-branch architecture that provides convolutional encoding of multiple feature resolutions, inspired by HRNet [73]. Last, we proposed an efficient fine-grained search strategy to train HR-NAS, which effectively explores the search space, and finds optimal architectures given various tasks and computation resources. As shown in Fig. 1 (a), HR-NAS is capable of achieving state-of-the-art trade-offs between performance and FLOPs for three dense prediction tasks and an image classification task, given only small computational budgets. For example, HR-NAS surpasses SqueezeNAS [63] that is specially designed for semantic segmentation while improving efficiency by 45.9%. Code is available at https://github.com/dingmyu/HR-NAS.

46 citations

Journal ArticleDOI
TL;DR: The hypothesis that two segregate systems process navigational memory for large-scale environments and spatial memory in small- scale environments is supported.
Abstract: Recent reports show that humans and animals do not acquire information about routes and object locations in the same way. In spatial memory, a specific sub-system is hypothesized to be involved in encoding, storing and recalling navigational information, and it is segregated from the sub-system devoted to small-scale environment. We assessed this hypothesis in a sample of patients treated surgically for intractable temporal lobe epilepsy. We found double dissociations between learning and recall of spatial positions in large space versus small space. These results strongly support the hypothesis that two segregate systems process navigational memory for large-scale environments and spatial memory in small-scale environments.

46 citations

Journal ArticleDOI
TL;DR: Working memory for ASL is sensitive to irrelevant signed input (and other structured visual input) in a manner similar to the effects of irrelevant auditory input on working memory for speech, suggesting parallels to visuospatial working memory.
Abstract: We report results showing that working memory for American Sign Language (ASL) is sensitive to irrelevant signed input (and other structured visual input) in a manner similar to the effects of irrelevant auditory input on working memory for speech. Deaf signers were disrupted on serial recall of lists of ASL signs when either pseudosigns or moving shapes were presented during a retention interval. Hearing subjects asked to recall lists of printed English words did not show disruption under the same interference conditions. The results favor models that hypothesize modality-specific representations of language within working memory, as opposed to amodal representations. The results further indicate that working memory for sign language involves visual or quasi-visual representations, suggesting parallels to visuospatial working memory.

46 citations

Journal ArticleDOI
TL;DR: In this paper, the authors examined whether spatial location information is more likely to be encoded with the memory representation of objects than of words and found that different processes are involved in encoding item and location information for words but not for objects.
Abstract: Four experiments examine whether spatial location information is more likely to be encoded with the memory representation of objects than of words. Sixteen objects or the one-word verbal labels for each were studied on a matrix display, followed by a recall test and then a relocation test. In each experiment, an independent variable known to affect item recall was introduced to test whether spatial location memory would concern itantly vary for both objects and words. In Experiment 1, recall of both objects and words increased with age of the subjects. However, relocation accuracy increased for objects but not for words. In Experiment 2, visual imagery instructions generally improved memory for words without affecting relocation accuracy. In Experiments 3 and 4, prolonging the test delay diminished recall for objects and words. However, relocation accuracy decreased only for the objects. In each experiment, item memory was affected independently of location memory for words but not for objects. The results suggest that different processes are involved in encoding item and location information for words but not for objects.

45 citations

Journal ArticleDOI
TL;DR: This paper proposes centralized uncoded placement and linear delivery schemes which are optimized by solving a linear program and derives a lower bound on the delivery memory tradeoff with uncoded placed that accounts for the heterogeneity in cache sizes.
Abstract: In cache-aided networks, the server populates the cache memories at the users during low-traffic periods in order to reduce the delivery load during peak-traffic hours. In turn, there exists a fundamental tradeoff between the delivery load on the server and the cache sizes at the users. In this paper, we study this tradeoff in a multicast network, where the server is connected to users with unequal cache sizes and the number of users is less than or equal to the number of library files. We propose centralized uncoded placement and linear delivery schemes which are optimized by solving a linear program. Additionally, we derive a lower bound on the delivery memory tradeoff with uncoded placement that accounts for the heterogeneity in cache sizes. We explicitly characterize this tradeoff for the case of three end-users, as well as an arbitrary number of end-users when the total memory size at the users is small, and when it is large. Next, we consider a system where the server is connected to the users via rate-limited links of different capacities and the server assigns the users’ cache sizes subject to a total cache budget. We characterize the optimal cache sizes that minimize the delivery completion time with uncoded placement and linear delivery. In particular, the optimal memory allocation balances between assigning larger cache sizes to users with low capacity links and uniform memory allocation.

45 citations


Network Information
Related Topics (5)
Artificial neural network
207K papers, 4.5M citations
83% related
Deep learning
79.8K papers, 2.1M citations
83% related
Feature extraction
111.8K papers, 2.1M citations
82% related
Convolutional neural network
74.7K papers, 2M citations
81% related
Cluster analysis
146.5K papers, 2.9M citations
81% related
Performance
Metrics
No. of papers in the topic in previous years
YearPapers
20231,083
20222,253
2021450
2020378
2019358
2018363