Subscribe to the gold package and get unlimited access to Shamra Academy
Register a new userDeepHash: Getting Regularization, Depth and Fine-Tuning Right
89
0
0.0
(
0
)
Added by
Vijay Chandrasekhar
Publication date
2015
fields
Informatics Engineering
and research's language is
English
Ask ChatGPT about the research
No Arabic abstract
This work focuses on representing very high-dimensional global image descriptors using very compact 64-1024 bit binary hashes for instance retrieval. We propose DeepHash: a hashing scheme based on deep networks. Key to making DeepHash work at extremely low bitrates are three important considerations -- regularization, depth and fine-tuning -- each requiring solutions specific to the hashing problem. In-depth evaluation shows that our scheme consistently outperforms state-of-the-art methods across all data sets for both Fisher Vectors and Deep Convolutional Neural Network features, by up to 20 percent over other schemes. The retrieval performance with 256-bit hashes is close to that of the uncompressed floating point features -- a remarkable 512 times compression.
rate research
Read More
Fine-tuning pre-trained cross-lingual language models can transfer task-specific supervision from one language to the others. In this work, we propose to improve cross-lingual fine-tuning with consistency regularization. Specifically, we use example consistency regularization to penalize the prediction sensitivity to four types of data augmentations, i.e., subword sampling, Gaussian noise, code-switch substitution, and machine translation. In addition, we employ model consistency to regularize the models trained with two augment
Fine-tuning pre-trained language models such as BERT has become a common practice dominating leaderboards across various NLP tasks. Despite its recent success and wide adoption, this process is unstable when there are only a small number of training samples available. The brittleness of this process is often reflected by the sensitivity to random seeds. In this paper, we propose to tackle this problem based on the noise stability property of deep nets, which is investigated in recent literature (Arora et al., 2018; Sanyal et al., 2020). Specifically, we introduce a novel and effective regularization method to improve fine-tuning on NLP tasks, referred to as Layer-wise Noise Stability Regularization (LNSR). We extend the theories about adding noise to the input and prove that our method gives a stabler regularization effect. We provide supportive evidence by experimentally confirming that well-performing models show a low sensitivity to noise and fine-tuning with LNSR exhibits clearly higher generalizability and stability. Furthermore, our method also demonstrates advantages over other state-of-the-art algorithms including L2-SP (Li et al., 2018), Mixout (Lee et al., 2020) and SMART (Jiang et al., 2020).
Multi-threaded programs have traditionally fallen into one of two domains: cooperative and competitive. These two domains have traditionally remained mostly disjoint, with cooperative threading used for increasing throughput in compute-intensive applications such as scientific workloads and cooperative threading used for increasing responsiveness in interactive applications such as GUIs and games. As multicore hardware becomes increasingly mainstream, there is a need for bridging these two disjoint worlds, because many applications mix interaction and computation and would benefit from both cooperative and competitive threading. In this paper, we present techniques for programming and reasoning about parallel interactive applications that can use both cooperative and competitive threading. Our techniques enable the programmer to write rich parallel interactive programs by creating and synchronizing with threads as needed, and by assigning threads user-defined and partially ordered priorities. To ensure important responsiveness properties, we present a modal type system analogous to S4 modal logic that precludes low-priority threads from delaying high-priority threads, thereby statically preventing a crucial set of priority-inversion bugs. We then present a cost model that allows reasoning about responsiveness and completion time of well-typed programs. The cost model extends the traditional work-span model for cooperative threading to account for competitive scheduling decisions needed to ensure responsiveness. Finally, we show that our proposed techniques are realistic by implementing them as an extension to the Standard ML language.
Pre-trained language models (LM) have become go-to text representation encoders. Prior research used deep LMs to encode text sequences such as sentences and passages into single dense vector representations. These dense representations have been used in efficient text comparison and embedding-based retrieval. However, dense encoders suffer in low resource situations. Many techniques have been developed to solve this problem. Despite their success, not much is known about why this happens. This paper shows that one cause lies in the readiness of the LM to expose its knowledge through dense representation in fine-tuning, which we term Optimization Readiness. To validate the theory, we present Condenser, a general pre-training architecture based on Transformer LMs, to improve dense optimization readiness. We show that fine-tuning from Condenser significantly improves performance for small and/or noisy training sets.
Fine-tuning from pre-trained ImageNet models has become the de-facto standard for various computer vision tasks. Current practices for fine-tuning typically involve selecting an ad-hoc choice of hyperparameters and keeping them fixed to values normally used for training from scratch. This paper re-examines several common practices of setting hyperparameters for fine-tuning. Our findings are based on extensive empirical evaluation for fine-tuning on various transfer learning benchmarks. (1) While prior works have thoroughly investigated learning rate and batch size, momentum for fine-tuning is a relatively unexplored parameter. We find that the value of momentum also affects fine-tuning performance and connect it with previous theoretical findings. (2) Optimal hyperparameters for fine-tuning, in particular, the effective learning rate, are not only dataset dependent but also sensitive to the similarity between the source domain and target domain. This is in contrast to hyperparameters for training from scratch. (3) Reference-based regularization that keeps models close to the initial model does not necessarily apply for dissimilar datasets. Our findings challenge common practices of fine-tuning and encourages deep learning practitioners to rethink the hyperparameters for fine-tuning.
Log in to be able to interact and post comments
comments
Fetching comments
Sign in to be able to follow your search criteria
