There are 4,825 ham and 747 spam messages. This indicates the data is imbalanced which needs to be fixed. The top ham message is "Sorry, I'll call later", whereas the top spam message is "Please call our customer service…" which occurred 30 and 4 times, respectively. First, let's create a separate dataframe for ham and spam messages and convert it to NumPy array and then to a list to generate WordCloud later. Since it is a text data, there are many unnecessary stopwords like articles, prepositions etc., which needs to be removed from the data.

Lee Sedol, a world-class Go Champion, was flummoxed by the 37th move Deepmind's AlphaGo made in the second match of the famous 2016 series. So flummoxed that it took him nearly 15 minutes to formulate a response. The move was strange to other experienced Go players as well, with one commentator suggesting it was a mistake. In fact, it was a canonical example of an artificial intelligence algorithm learning something that seemed to go beyond just pattern recognition in data -- learning something strategic and even creative. Indeed, beyond just feeding the algorithm past examples of Go champions playing games, Deepmind developers trained AlphaGo by having it play many millions of matches against itself.

In this series around GPT language model, we will focus on the paper "Extract Training Data from Large Language Models" The authors want to show that they can extract verbatim data from a language model such as GPT-2. More interestingly, they explain that they can extract verbatim that have appeared only a few times in the training data from the model itself. Naturally, that can be very dangerous if you own a company and you are using customers' data to train a language model. In their own words, "the paper demonstrates that (…), an adversary can perform a training data extraction attack to recover individual training examples by querying the language model." Who would want to risk leaking private information?

Researchers from all over the world contribute to this repository as a prelude to the peer review process for publication in traditional journals. The articles listed below represent a small fraction of all articles appearing on the preprint server. They are listed in no particular order with a link to each paper along with a brief overview. Links to GitHub repos are provided when available. Especially relevant articles are marked with a "thumbs up" icon.

Description The artificial intelligence is a large field includes many techniques to make machine thinks. Therefore, in this course, we investigate the mimicking of human intelligence on machines by introducing a modern algorithm of artificial intelligence named convolutional neural network which is a technique of deep learning for computers to make the machine learn and expert. In this course, we present an overview of deep learning in which, we introduce the notion and classification of convolutional neural networks. We gives also the definition and the advantages of CNNs. In this course, we provide the tricks to elaborate your own architecture of CNN and the hardware and software to design a CNN model. In the end, we present the limitation and future challenges of CNN.

While cloud computing continues to gain favor, only a limited number of companies have embraced machine learning based in the cloud. Hive wants to change this by allowing enterprises to access hosted machine learning models via APIs. Hive has had particular success in the area of content moderation, thanks to its deep learning models that help companies interpret unstructured data, like images, videos, and audio. But it's also expanding into areas like advertising and sponsorship measurement as it seeks to find other areas that would benefit from intelligent automation. In an interview with VentureBeat, Hive CEO Kevin Guo said the company kept relatively quiet as it sought to prove its models work.

In statistical learning theory, there is something called the VC Dimension of an algorithm (https://en.m.wikipedia.org/wiki/Vapnik%E2%80%93Chervonenkis_dimension) - the VC dimension is apparently what describes the relartive level of complexity a machine learning algorithm can capture. Does this concept of VC Dimension carry over to models in time series analysis? Is it possible to show that LSTM's have a higher VC dimension compared to ARIMA style models? Supposedly, neural network based time series models were developed because modeols like ARIMA was unable to provide reliable estimates for bigger and complex datasets. Mathematically speaking, what allows a LSTM to capture more variation and complexity in a dataset compared to ARIMA?

This paper proposes a Fully Spiking Hybrid Neural Network (FSHNN) for energy-efficient and robust object detection in resource-constrained platforms. The network architecture is based on Convolutional SNN using leaky-integrate-fire neuron models. The model combines unsupervised Spike Time-Dependent Plasticity (STDP) learning with back-propagation (STBP) learning methods and also uses Monte Carlo Dropout to get an estimate of the uncertainty error. FSHNN provides better accuracy compared to DNN based object detectors while being 150X energy-efficient. It also outperforms these object detectors, when subjected to noisy input data and less labeled training data with a lower uncertainty error.

Motivation: The activity of the adaptive immune system is governed by T-cells and their specific T-cell receptors (TCR), which selectively recognize foreign antigens. Recent advances in experimental techniques have enabled sequencing of TCRs and their antigenic targets (epitopes), allowing to research the missing link between TCR sequence and epitope binding specificity. Scarcity of data and a large sequence space make this task challenging, and to date only models limited to a small set of epitopes have achieved good performance. Here, we establish a k-nearest-neighbor (K-NN) classifier as a strong baseline and then propose TITAN (Tcr epITope bimodal Attention Networks), a bimodal neural network that explicitly encodes both TCR sequences and epitopes to enable the independent study of generalization capabilities to unseen TCRs and/or epitopes. Results: By encoding epitopes at the atomic level with SMILES sequences, we leverage transfer learning and data augmentation to enrich the input data space and boost performance. TITAN achieves high performance in the prediction of specificity of unseen TCRs (ROC-AUC 0.87 in 10-fold CV) and surpasses the results of the current state-of-the-art (ImRex) by a large margin. Notably, our Levenshtein-distance-based K-NN classifier also exhibits competitive performance on unseen TCRs. While the generalization to unseen epitopes remains challenging, we report two major breakthroughs. First, by dissecting the attention heatmaps, we demonstrate that the sparsity of available epitope data favors an implicit treatment of epitopes as classes. This may be a general problem that limits unseen epitope performance for sufficiently complex models. Second, we show that TITAN nevertheless exhibits significantly improved performance on unseen epitopes and is capable of focusing attention on chemically meaningful molecular structures.

Reinforcement learning techniques achieved human-level performance in several tasks in the last decade. However, in recent years, the need for interpretability emerged: we want to be able to understand how a system works and the reasons behind its decisions. Not only we need interpretability to assess the safety of the produced systems, we also need it to extract knowledge about unknown problems. While some techniques that optimize decision trees for reinforcement learning do exist, they usually employ greedy algorithms or they do not exploit the rewards given by the environment. This means that these techniques may easily get stuck in local optima. In this work, we propose a novel approach to interpretable reinforcement learning that uses decision trees. We present a two-level optimization scheme that combines the advantages of evolutionary algorithms with the advantages of Q-learning. This way we decompose the problem into two sub-problems: the problem of finding a meaningful and useful decomposition of the state space, and the problem of associating an action to each state. We test the proposed method on three well-known reinforcement learning benchmarks, on which it results competitive with respect to the state-of-the-art in both performance and interpretability. Finally, we perform an ablation study that confirms that using the two-level optimization scheme gives a boost in performance in non-trivial environments with respect to a one-layer optimization technique.