Subscribe to the gold package and get unlimited access to Shamra Academy
Register a new userSub-Linear Memory: How to Make Performers SLiM
85
0
0.0
(
0
)
Added by
Valerii Likhosherstov
Publication date
2020
fields
Informatics Engineering
and research's language is
English
Ask ChatGPT about the research
No Arabic abstract
The Transformer architecture has revolutionized deep learning on sequential data, becoming ubiquitous in state-of-the-art solutions for a wide variety of applications. Yet vanilla Transformers are notoriously resource-expensive, requiring $O(L^2)$ in serial time and memory as functions of input length $L$. Recent works proposed various linear self-attention mechanisms, scaling only as $O(L)$ for serial computation. We perform a thorough analysis of recent Transformer mechanisms with linear self-attention, Performers, in terms of overall computational complexity. We observe a remarkable computational flexibility: forward and backward propagation can be performed with no approximations using sublinear memory as a function of $L$ (in addition to negligible storage for the input sequence), at a cost of greater time complexity in the parallel setting. In the extreme case, a Performer consumes only $O(1)$ memory during training, and still requires $O(L)$ time. This discovered time-memory tradeoff can be used for training or, due to complete backward-compatibility, for fine-tuning on a low-memory device, e.g. a smartphone or an earlier-generation GPU, thus contributing towards decentralized and democratized deep learning.
rate research
Read More
This is a story about making quantum computers speak, and doing so in a quantum-native, compositional and meaning-aware manner. Recently we did question-answering with an actual quantum computer. We explain what we did, stress that this was all done in terms of pictures, and provide many pointers to the related literature. In fact, besides natural language, many other things can be implemented in a quantum-native, compositional and meaning-aware manner, and we provide the reader with some indications of that broader pictorial landscape, including our account on the notion of compositionality. We also provide some guidance for the actual execution, so that the reader can give it a go as well.
Several candidates for accreting magnetars have been proposed recently by different authors. Existence of such systems contradicts the standard magnetic field decay scenario where a large magnetic field of a neutron star reaches $lesssim$ few$times 10^{13}$G at ages $gtrsim 1$ Myr. Among other sources, the high mass X-ray binary 4U0114+65 seems to have a strong magnetic field around $10^{14}$ G. We develop a new Bayesian estimate for the kinematic age and demonstrate that 4U0114+65 has kinematic age 2.4-5 Myr ($95%$ credential interval) since the formation of the neutron star. We discuss which conditions are necessary to explain the potential existence of magnetars in accreting high-mass binaries with ages about few Myrs and larger. Three necessary ingredients are: the Hall attractor to prevent rapid decay of dipolar field, relatively rapid cooling of the crust in order to avoid Ohmic decay due to phonons, and finally, low values of the parameter $Q$ to obtain long Ohmic time scale due to impurities. If age and magnetic field estimates for proposed accreting magnetars are correct, then these systems set the strongest limit on the crust impurity for a selected sample of neutron stars and provide evidence in favour of the Hall attractor.
New text as data techniques offer a great promise: the ability to inductively discover measures that are useful for testing social science theories of interest from large collections of text. We introduce a conceptual framework for making causal inferences with discovered measures as a treatment or outcome. Our framework enables researchers to discover high-dimensional textual interventions and estimate the ways that observed treatments affect text-based outcomes. We argue that nearly all text-based causal inferences depend upon a latent representation of the text and we provide a framework to learn the latent representation. But estimating this latent representation, we show, creates new risks: we may introduce an identification problem or overfit. To address these risks we describe a split-sample framework and apply it to estimate causal effects from an experiment on immigration attitudes and a study on bureaucratic response. Our work provides a rigorous foundation for text-based causal inferences.
We have been working on speech synthesis for rakugo (a traditional Japanese form of verbal entertainment similar to one-person stand-up comedy) toward speech synthesis that authentically entertains audiences. In this paper, we propose a novel evaluation methodology using synthesized rakugo speech and real rakugo speech uttered by professional performers of three different ranks. The naturalness of the synthesized speech was comparable to that of the human speech, but the synthesized speech entertained listeners less than the performers of any rank. However, we obtained some interesting insights into challenges to be solved in order to achieve a truly entertaining rakugo synthesizer. For example, naturalness was not the most important factor, even though it has generally been emphasized as the most important point to be evaluated in the conventional speech synthesis field. More important factors were the understandability of the content and distinguishability of the characters in the rakugo story, both of which the synthesized rakugo speech was relatively inferior at as compared with the professional performers. We also found that fundamental frequency fo modeling should be further improved to better entertain audiences. These results show important steps to reaching authentically entertaining speech synthesis.
We introduce Performers, Transformer architectures which can estimate regular (softmax) full-rank-attention Transformers with provable accuracy, but using only linear (as opposed to quadratic) space and time complexity, without relying on any priors such as sparsity or low-rankness. To approximate softmax attention-kernels, Performers use a novel Fast Attention Via positive Orthogonal Random features approach (FAVOR+), which may be of independent interest for scalable kernel methods. FAVOR+ can be also used to efficiently model kernelizable attention mechanisms beyond softmax. This representational power is crucial to accurately compare softmax with other kernels for the first time on large-scale tasks, beyond the reach of regular Transformers, and investigate optimal attention-kernels. Performers are linear architectures fully compatible with regular Transformers and with strong theoretical guarantees: unbiased or nearly-unbiased estimation of the attention matrix, uniform convergence and low estimation variance. We tested Performers on a rich set of tasks stretching from pixel-prediction through text models to protein sequence modeling. We demonstrate competitive results with other examined efficient sparse and dense attention methods, showcasing effectiveness of the novel attention-learning paradigm leveraged by Performers.
suggested questions
Log in to be able to interact and post comments
comments
Fetching comments
Sign in to be able to follow your search criteria
