Hypernetworks are meta neural networks that generate weights for a main neural network in an end-to-end differentiable manner. Despite extensive applications ranging from multi-task learning to Bayesian deep learning, the problem of optimizing hypernetworks has not been studied to date. We observe that classical weight initialization methods like Glorot & Bengio (2010) and He et al. (2015), when applied directly on a hypernet, fail to produce weights for the mainnet in the correct scale. We develop principled techniques for weight initialization in hypernets, and show that they lead to more stable mainnet weights, lower training loss, and faster convergence. Meta-learning describes a broad family of techniques in machine learning that deals with the problem of learning to learn. An emerging branch of meta-learning involves the use of hypernetworks, which are meta neural networks that generate the weights of a main neural network to solve a given task in an end-to-end differentiable manner. Hypernetworks were originally introduced by Ha et al. (2016) as a way to induce weight-sharing and achieve model compression by training the same meta network to learn the weights belonging to different layers in the main network.

Blockchain security is becoming increasingly relevant in today's cyberspace as it extends its influence in many industries. This paper focuses on protecting the lowest level layer in the blockchain, particularly the P2P network that allows the nodes to communicate and share information. The P2P network layer may be vulnerable to several families of attacks, such as Distributed Denial of Service (DDoS), eclipse attacks, or Sybil attacks. This layer is prone to threats inherited from traditional P2P networks, and it must be analyzed and understood by collecting data and extracting insights from the network behavior to reduce those risks. We introduce Tikuna, an open-source tool for monitoring and detecting potential attacks on the Ethereum blockchain P2P network at an early stage. Tikuna employs an unsupervised Long Short Term Memory (LSTM) method based on Recurrent Neural Network (RNNs) to detect attacks and alert users. Empirical results indicate that the proposed approach significantly improves detection performance, with the ability to detect and classify attacks, including eclipse attacks, Covert Flash attacks, and others that target the Ethereum blockchain P2P network layer, with high accuracy. Our research findings demonstrate that Tikuna is a valuable security tool for assisting operators to efficiently monitor and safeguard the status of Ethereum validators and the wider P2P network. Keywords: Ethereum blockchain, security, P2P network, deep learning, anomaly detection, vulnerabilities, eclipse attacks.

As usual, we are holding this AMA on the regular basis with our CEO, Mr Oven Tao. Since we have some deliverables and that we would like to make use of this AMA to announce we have achieved and particularly for the topics that interest most of the Community members. Then the second part we will go on to the question and answer sections we have for this month. We have over 90 questions from our community. Some are I think the reason that we have this many questions. There was some newcomers and they raised many questions which are they can find from our website and but still I am select some of them and then to be asked in this AMA. And in this AMA as usual we have questions and for our CEO Owen to address. So let's start with the first session which is about the progress that we have made according to our road map. Owen are you ready to start?

An average transaction takes over an hour to verify. If security should be an issue for crypto assets, then there is no point investing in them. PoW of Bitcoin is nothing but meaningless hash computing that creates no real value for society. According to statistics, 70% of the world's computing power is consumed to mine cryptos, which is a million times Google's computing power. Bitcoin mining also consumed more power each year than 260 countries of the world, creating a huge waste.

In video segmentation, generating temporally consistent results across frames is as important as achieving frame-wise accuracy. Existing methods rely either on optical flow regularization or fine-tuning with test data to attain temporal consistency. However, optical flow is not always avail-able and reliable. Besides, it is expensive to compute. Fine-tuning the original model in test time is cost sensitive. This paper presents an efficient, intuitive, and unsupervised online adaptation method, AuxAdapt, for improving the temporal consistency of most neural network models. It does not require optical flow and only takes one pass of the video. Since inconsistency mainly arises from the model's uncertainty in its output, we propose an adaptation scheme where the model learns from its own segmentation decisions as it streams a video, which allows producing more confident and temporally consistent labeling for similarly-looking pixels across frames. For stability and efficiency, we leverage a small auxiliary segmentation network (AuxNet) to assist with this adaptation. More specifically, AuxNet readjusts the decision of the original segmentation network (Main-Net) by adding its own estimations to that of MainNet. At every frame, only AuxNet is updated via back-propagation while keeping MainNet fixed. We extensively evaluate our test-time adaptation approach on standard video benchmarks, including Cityscapes, CamVid, and KITTI. The results demonstrate that our approach provides label-wise accurate, temporally consistent, and computationally efficient adaptation (5+ folds overhead reduction comparing to state-of-the-art test-time adaptation methods).

London March 10, 2021 – Using blockchain to democratize machine learning, GNY today launched the world's first blockchain providing access to powerful machine learning functionalities on a secure and collaborative decentralized platform. Empowering businesses and organizations to solve large and complex problems, GNY's Mainnet will enable users to utilize its blockchain architecture to run machine learning analysis of any type of data – from financial records to public health information – without risking sensitive data theft. The Mainnet launches with 5 machine learning use cases, which will expand to dozens by the end of 2021 and eventually hundreds. The launch comes after 18 months of development and rigorous testing. To demonstrate its potential impact, one GNY test revealed how researchers could use Mainnet to set up a private blockchain to run comparative analysis of daily COVID-19 mortalities from individual cities.

As the performance and popularity of deep neural networks has increased, so too has their computational cost. There are many effective techniques for reducing a network's computational footprint (quantisation, pruning, knowledge distillation), but these lead to models whose computational cost is the same regardless of their input. Our human reaction times vary with the complexity of the tasks we perform: easier tasks (e.g. telling apart dogs from boat) are executed much faster than harder ones (e.g. telling apart two similar looking breeds of dogs). Driven by this observation, we develop a method for adaptive network complexity by attaching a small classification layer, which we call SideNet, to a large pretrained network, which we call MainNet. Given an input, the SideNet returns a classification if its confidence level, obtained via softmax, surpasses a user determined threshold, and only passes it along to the large MainNet for further processing if its confidence is too low. This allows us to flexibly trade off the network's performance with its computational cost. Experimental results show that simple single hidden layer perceptron SideNets added onto pretrained ResNet and BERT MainNets allow for substantial decreases in compute with minimal drops in performance on image and text classification tasks. We also highlight three other desirable properties of our method, namely that the classifications obtained by SideNets are calibrated, complementary to other compute reduction techniques, and that they enable the easy exploration of compute accuracy space.