Deep learning (DL) has been widely adopted those last years but they are computing-intensive method. Therefore, scientists proposed diverse optimization to accelerate their predictions for end-user applications. However, no single inference framework currently dominates in terms of performance. This paper takes a holistic approach to conduct an empirical comparison and analysis of four representative DL inference frameworks. First, given a selection of CPU-GPU configurations, we show that for a specific DL framework, different configurations of its settings may have a significant impact on the prediction speed, memory, and computing power. Second, to the best of our knowledge, this study is the first to identify the opportunities for accelerating the ensemble of co-localized models in the same GPU. This measurement study provides an in-depth empirical comparison and analysis of four representative DL frameworks and offers practical guidance for service providers to deploy and deliver DL predictions.

Contrastive learning has been demonstrated to be effective in enhancing pre-trained language models (PLMs) to derive superior universal sentence embeddings. However, existing contrastive methods still have two limitations. Firstly, previous works may acquire poor performance under domain shift settings, thus hindering the application of sentence representations in practice. We attribute this low performance to the over-parameterization of PLMs with millions of parameters. To alleviate it, we propose PromCSE (Prompt-based Contrastive Learning for Sentence Embeddings), which only trains small-scale \emph{Soft Prompt} (i.e., a set of trainable vectors) while keeping PLMs fixed. Secondly, the commonly used NT-Xent loss function of contrastive learning does not fully exploit hard negatives in supervised learning settings. To this end, we propose to integrate an Energy-based Hinge loss to enhance the pairwise discriminative power, inspired by the connection between the NT-Xent loss and the Energy-based Learning paradigm. Empirical results on seven standard semantic textual similarity (STS) tasks and a domain-shifted STS task both show the effectiveness of our method compared with the current state-of-the-art sentence embedding models. Our code is publicly avaliable at https://github.com/YJiangcm/PromCSE

Randomization is currently a widely used approach in Sim2Real transfer for data-driven learning algorithms in robotics. Still, most Sim2Real studies report results for a specific randomization technique and often on a highly customized robotic system, making it difficult to evaluate different randomization approaches systematically. To address this problem, we define an easy-to-reproduce experimental setup for a robotic reach-and-balance manipulator task, which can serve as a benchmark for comparison. We compare four randomization strategies with three randomized parameters both in simulation and on a real robot. Our results show that more randomization helps in Sim2Real transfer, yet it can also harm the ability of the algorithm to find a good policy in simulation. Fully randomized simulations and fine-tuning show differentiated results and translate better to the real robot than the other approaches tested.

Criminals have become increasingly experienced in using cryptocurrencies, such as Bitcoin, for money laundering. The use of cryptocurrencies can hide criminal identities and transfer hundreds of millions of dollars of dirty funds through their criminal digital wallets. However, this is considered a paradox because cryptocurrencies are goldmines for open-source intelligence, giving law enforcement agencies more power when conducting forensic analyses. This paper proposed Inspection-L, a graph neural network (GNN) framework based on a self-supervised Deep Graph Infomax (DGI) and Graph Isomorphism Network (GIN), with supervised learning algorithms, namely Random Forest (RF), to detect illicit transactions for anti-money laundering (AML). To the best of our knowledge, our proposal is the first to apply self-supervised GNNs to the problem of AML in Bitcoin. The proposed method was evaluated on the Elliptic dataset and shows that our approach outperforms the state-of-the-art in terms of key classification metrics, which demonstrates the potential of self-supervised GNN in the detection of illicit cryptocurrency transactions.

Dublin has long been home to Big Tech's European outposts, drawn by low taxes and Ireland's position as the only English-speaking country in the European Union. Historically, however, this has negatively affected local startups: Big salaries and cushy positions at Big Tech companies made it difficult for smaller, nimbler companies to compete. That situation is finally changing: "Over the past few years, the culture has shifted away from Big Tech," says Nicola McClafferty, chair of the Irish Venture Capital Association and a partner in Molten Ventures, a venture capital firm operating in Ireland. "We're seeing more and more people and talent wanting to come out of those companies, and really thinking about joining earlier-stage and high-growth startups." That's in part down to Irish startup successes like communications platform Intercom and payments system Stripe, which have proven homegrown wins are possible.

Since 1995, the FDA has authorized more than 500 AI/ML-enabled medical devices via 510(k) clearance, granted De Novo request, or approved PMA. This week the FDA published an updated list with 178 new devices that were authorized through July 2022. According to the FDA, their list is based on publicly available information and is not a comprehensive resource of FDA approved AI/ML-enabled medical devices. In today's DeepTech newsletter I'm sharing a high level analysis of the 521 devices on the list, charts to visualize the data, and a summary of milestones. Note: According to the FDA their list is based on publicly available information and is not a comprehensive resource of approved AI/ML-enabled medical devices.

For two decades, the formalism of quantum mechanics has been successfully used to describe human decision processes, situations of heuristic reasoning, and the contextuality of concepts and their combinations. The phenomenon of 'categorical perception' has put us on track to find a possible deeper cause of the presence of this quantum structure in human cognition. Thus, we show that in an archetype of human perception consisting of the reconciliation of a bottom up stimulus with a top down cognitive expectation pattern, there arises the typical warping of categorical perception, where groups of stimuli clump together to form quanta, which move away from each other and lead to a discretization of a dimension. The individual concepts, which are these quanta, can be modeled by a quantum prototype theory with the square of the absolute value of a corresponding Schr\"odinger wave function as the fuzzy prototype structure, and the superposition of two such wave functions accounts for the interference pattern that occurs when these concepts are combined. Using a simple quantum measurement model, we analyze this archetype of human perception, provide an overview of the experimental evidence base for categorical perception with the phenomenon of warping leading to quantization, and illustrate our analyses with two examples worked out in detail.

It has been known in the mathematical community In this paper we use a simple and naive approach since 1897 --- although only clearly since for converting vocabulary words into vectors: use (Hensel, 1918) -- that there is an unusual and unexpected whatever the unicode bit sequence for the word family of distance metrics based on prime would be; this bit sequence can also be viewed as an numbers which can be used instead of Euclidean integer vector with one element. This is of course metrics, which have infinitesimals (to support calculus), extremely arbitrary and subject to the whims of the triangle inequality (to support geometry), the unicode consortium, but it is the most common and other useful properties all the while maintaining way to represent text from any human language on mathematical consistency. They are known a computer.

In this study, we synthesize a novel dynamical approach for ant colonies enabling them to migrate to new nest sites in a self-organizing fashion. In other words, we realize ant colony migration as a self-organizing phenotype-level collective behavior. For this purpose, we first segment the edges of the graph of ants' pathways. Then, each segment, attributed to its own pheromone profile, may host an ant. So, multiple ants may occupy an edge at the same time. Thanks to this segment-wise edge formulation, ants have more selection options in the course of their pathway determination, thereby increasing the diversity of their colony's emergent behaviors. In light of the continuous pheromone dynamics of segments, each edge owns a spatio-temporal piece-wise continuous pheromone profile in which both deposit and evaporation processes are unified. The passive dynamics of the proposed migration mechanism is sufficiently rich so that an ant colony can migrate to the vicinity of a new nest site in a self-organizing manner without any external supervision. In particular, we perform extensive simulations to test our migration dynamics applied to a colony including 500 ants traversing a pathway graph comprising 200 nodes and 4000 edges which are segmented based on various resolutions. The obtained results exhibit the effectiveness of our strategy.

Word alignment which aims to extract lexicon translation equivalents between source and target sentences, serves as a fundamental tool for natural language processing. Recent studies in this area have yielded substantial improvements by generating alignments from contextualized embeddings of the pre-trained multilingual language models. However, we find that the existing approaches capture few interactions between the input sentence pairs, which degrades the word alignment quality severely, especially for the ambiguous words in the monolingual context. To remedy this problem, we propose Cross-Align to model deep interactions between the input sentence pairs, in which the source and target sentences are encoded separately with the shared self-attention modules in the shallow layers, while cross-lingual interactions are explicitly constructed by the cross-attention modules in the upper layers. Besides, to train our model effectively, we propose a two-stage training framework, where the model is trained with a simple Translation Language Modeling (TLM) objective in the first stage and then finetuned with a self-supervised alignment objective in the second stage. Experiments show that the proposed Cross-Align achieves the state-of-the-art (SOTA) performance on four out of five language pairs.