The Prentice Hall Essence of Computing Series provides a concise, practical and uniform introduction to the core components of an undergraduate computer science degree. Acknowledging the recent changes within Higher Education, this approach uses a variety of pedagogical tools - case studies, worked examples and self-test questions, to underpin the student's learning. The Essence of Artificial Intelligence provides a concise and accessible introduction to the topic for students with no prior knowledge of AI. Taking a pragmatic approach to the subject, this book de-mystifies and makes AI concrete and transparent. Examples and Algorithms are given throughout and can be sensibly implemented in a range of different languages.

Google the "Tech Giant" and Coursera the "Leading Learning Firm" recently collaborated to produce a "Machine Learning specialization" course. These two have teamed up on numerous occasions and developed numerous courses for programmers and IT professionals earlier. This machine learning specialization course is a successor to the machine learning crash course which was an introduction to machine learning for beginner developers who wanted to explore the field of machine learning. This Specialization consists of five courses and has also given a good share to practical usage and applications. This new machine learning specialization course is called "Machine learning with TensorFlow on Google Cloud Platform", as it widely discusses and explains the usage of Google's TensorFlow which was on beta until 27th of April 2018.

During her freshman year, Stephanie Tena, a 16-year-old programmer, was searching the internet for coding programs and came across a website for an organization called AI4All, which runs an artificial-intelligence summer camp for high-schoolers. On the site, a group of girls her age were gathered around an autonomous car in front of the iconic arches of Stanford's campus. "AI will change the world," the text read. "Who will change AI?" How technology and globalization are changing what it means to work Read more Tena thought maybe she could. She lives in a trailer park in California's Central Valley; her mom, a Mexican immigrant from Michoacán, picks strawberries in the nearby fields.

I visited Beijing last week and learned a great deal from Microsoft's AI //innovate event in China - our 2nd largest developer base for Cognitive Services on Azure. The most heart-felt demo was Harry Shum could communicate with an almost deaf student of Nanjing University of Technology. Using two mobile phones with Microsoft Translator, the deaf student not only had no problem to communicate with Harry but also removed his language barriers between Chinese and English! I started speech recognition research as a graduate student in Beijing's Tsinghua University more than 35 years ago. My graduate student dream was to help people communicate better without language barriers.

This is a PyTorch implementation of Listen, Attend and Spell (LAS) published in ICASSP 2016 (Student Paper Award). Please feel free to use/modify them, any bug report or improvement suggestion will be appreciated. This implement achieves about 34% phoneme error rate on TIMIT's testing set (using original setting in the paper without hyper parameter tuning, models are stored in checkpoint/). It's not a remarkable score but please notice that deep end2end ASR without special designed loss function such as LAS requires larger corpus to achieve outstanding performance. Result of the first sample in TIMIT testing set.

We consider the transfer of experience samples (i.e., tuples < s, a, s', r >) in reinforcement learning (RL), collected from a set of source tasks to improve the learning process in a given target task. Most of the related approaches focus on selecting the most relevant source samples for solving the target task, but then all the transferred samples are used without considering anymore the discrepancies between the task models. In this paper, we propose a model-based technique that automatically estimates the relevance (importance weight) of each source sample for solving the target task. In the proposed approach, all the samples are transferred and used by a batch RL algorithm to solve the target task, but their contribution to the learning process is proportional to their importance weight. By extending the results for importance weighting provided in supervised learning literature, we develop a finite-sample analysis of the proposed batch RL algorithm. Furthermore, we empirically compare the proposed algorithm to state-of-the-art approaches, showing that it achieves better learning performance and is very robust to negative transfer, even when some source tasks are significantly different from the target task.

Despite all the impressive advances of recurrent neural networks, sequential data is still in need of better modelling. Truncated backpropagation through time (TBPTT), the learning algorithm most widely used in practice, suffers from the truncation bias, which drastically limits its ability to learn long-term dependencies.The Real-Time Recurrent Learning algorithm (RTRL) addresses this issue, but its high computational requirements make it infeasible in practice. The Unbiased Online Recurrent Optimization algorithm (UORO) approximates RTRL with a smaller runtime and memory cost, but with the disadvantage of obtaining noisy gradients that also limit its practical applicability. In this paper we propose the Kronecker Factored RTRL (KF-RTRL) algorithm that uses a Kronecker product decomposition to approximate the gradients for a large class of RNNs. We show that KF-RTRL is an unbiased and memory efficient online learning algorithm. Our theoretical analysis shows that, under reasonable assumptions, the noise introduced by our algorithm is not only stable over time but also asymptotically much smaller than the one of the UORO algorithm. We also confirm these theoretical results experimentally. Further, we show empirically that the KF-RTRL algorithm captures long-term dependencies and almost matches the performance of TBPTT on real world tasks by training Recurrent Highway Networks on a synthetic string memorization task and on the Penn TreeBank task, respectively. These results indicate that RTRL based approaches might be a promising future alternative to TBPTT.

Humans excel at continually acquiring and fine-tuning knowledge over sustained time spans. This ability, typically referred to as lifelong learning, is crucial for artificial agents interacting in real-world, dynamic environments where i) the number of tasks to be learned is not pre-defined, ii) training samples become progressively available over time, and iii) annotated samples may be very sparse. In this paper, we propose a dual-memory self-organizing system that learns spatiotemporal representations from videos. The architecture draws inspiration from the interplay of the hippocampal and neocortical systems in the mammalian brain argued to mediate the complementary tasks of quickly integrating specific experiences, i.e., episodic memory (EM), and slowly learning generalities from episodic events, i.e., semantic memory (SM). The complementary memories are modeled as recurrent self-organizing neural networks: The EM quickly adapts to incoming novel sensory observations via competitive Hebbian Learning, whereas the SM progressively learns compact representations by using task-relevant signals to regulate intrinsic levels of neurogenesis and neuroplasticity. For the consolidation of knowledge, trajectories of neural reactivations are periodically replayed to both networks. We analyze and evaluate the performance of our approach with the CORe50 benchmark dataset for continuous object recognition from videos. We show that the proposed approach significantly outperforms current (supervised) methods of lifelong learning in three different incremental learning scenarios, and that due to the unsupervised nature of neural network self-organization, our approach can be used in scenarios where sample annotations are sparse.

Knowledge tracing (KT) refers to a machine learning technique to assess a student's level of understanding (so-called knowledge state) of a certain concept based on the student performance on problem solving. KT accepts a series of question-answer pairs as an input and iteratively updates the knowledge state of the student, eventually returning the probability of the student solving an unseen question. From the viewpoint of neuroeducation (the field of applying neuroscience, cognitive science, and psychology to education), however, KT leaves much room for improvement in terms of explaining the complex process of human learning. In this paper, we identify three problems of KT (namely non adaptive knowledge growth, neglected latent information, and unintended negative influence) and propose a memory-network-based technique named intelligent knowledge tracing (IKT) to address them, thus approaching one step closer to understanding the complex mechanism underlying human learning. In addition, we propose a new performance metric called correct update count (CUC) that can measure the degree of unintended negative influence, thus quantifying how closely a student model resembles the human learning process. The proposed CUC metric can complement the area under the curve (AUC) metric, allowing us to evaluate competing models more effectively. According to our experiments using a widely used public benchmark, IKT significantly (over two times) outperformed the existing KT approaches in terms of CUC, while preserving the correctness behavior measured in AUC.

Few-shot learning has become essential for producing models that generalize from few examples. In this work, we identify that metric scaling and metric task conditioning are important to improve the performance of few-shot algorithms. Our analysis reveals that simple metric scaling completely changes the nature of few-shot algorithm parameter updates. Metric scaling provides improvements up to 14% in accuracy for certain metrics on the mini-Imagenet 5-way 5-shot classification task. We further propose a simple and effective way of conditioning a learner on the task sample set, resulting in learning a task-dependent metric space. Moreover, we propose and empirically test a practical end-to-end optimization procedure based on auxiliary task co-training to learn a task-dependent metric space. The resulting few-shot learning model based on the task-dependent scaled metric achieves state of the art on mini-Imagenet. We confirm these results on another few-shot dataset that we introduce in this paper based on CIFAR100.