Stanford Computer scientist Christopher Re discussed the changing software paradigm, Software 2.0. He told the University's Human-Centered AI group that focusing on neural network-building, and other low-level tasks such as tweaking hyper-parameters, is not really where engineers can make their most valuable efforts. Some AI researchers' practices are as tired as a Michael Bay movie, to hear Christopher Re tell it. Wednesday, Re, who is a Stanford University associate professor of computer science, gave a talk for the University's Human-Centered Artificial Intelligence institute. His topic: "Weird new things are happening in software."

TalkingData is a data intelligence service provider that offers data products and services to provide businesses insights on consumer behavior, preferences, and trends. One of TalkingData's core services is leveraging machine learning and deep learning models to predict consumer behaviors (e.g., likelihood of a particular group to buy a house or a car) and use these insights for targeted advertising. For example, a car dealer will only want to show their ads to customers who the model predicts are most likely to buy a car in the next three months. Initially, TalkingData was building an XGBoost model for these types of predictions, but their data science team wanted to explore whether deep learning models could have a significant performance improvement for their use case. After experimentation, their data scientists built a model on PyTorch, an open source deep learning framework, that achieved a 13% improvement on recall rate.

We study the benefits of complex-valued weights for neural networks. We prove that shallow complex neural networks with quadratic activations have no spurious local minima. In contrast, shallow real neural networks with quadratic activations have infinitely many spurious local minima under the same conditions. In addition, we provide specific examples to demonstrate that complex-valued weights turn poor local minima into saddle points. The activation function CReLU is also discussed to illustrate the superiority of analytic activations in complex-valued neural networks.

Rocket and MiniRocket, while two of the fastest methods for time series classification, are both somewhat less accurate than the current most accurate methods (namely, HIVE-COTE and its variants). We show that it is possible to significantly improve the accuracy of MiniRocket (and Rocket), with some additional computational expense, by expanding the set of features produced by the transform, making MultiRocket (for MiniRocket with Multiple Features) overall the single most accurate method on the datasets in the UCR archive, while still being orders of magnitude faster than any algorithm of comparable accuracy other than its precursors.

Several recent works have shown separation results between deep neural networks, and hypothesis classes with inferior approximation capacity such as shallow networks or kernel classes. On the other hand, the fact that deep networks can efficiently express a target function does not mean this target function can be learned efficiently by deep neural networks. In this work we study the intricate connection between learnability and approximation capacity. We show that learnability with deep networks of a target function depends on the ability of simpler classes to approximate the target. Specifically, we show that a necessary condition for a function to be learnable by gradient descent on deep neural networks is to be able to approximate the function, at least in a weak sense, with shallow neural networks. We also show that a class of functions can be learned by an efficient statistical query algorithm if and only if it can be approximated in a weak sense by some kernel class. We give several examples of functions which demonstrate depth separation, and conclude that they cannot be efficiently learned, even by a hypothesis class that can efficiently approximate them.

Probabilistic time series forecasting involves estimating the distribution of future based on its history, which is essential for risk management in downstream decision-making. We propose a deep state space model for probabilistic time series forecasting whereby the non-linear emission model and transition model are parameterized by networks and the dependency is modeled by recurrent neural nets. We take the automatic relevance determination (ARD) view and devise a network to exploit the exogenous variables in addition to time series. In particular, our ARD network can incorporate the uncertainty of the exogenous variables and eventually helps identify useful exogenous variables and suppress those irrelevant for forecasting. The distribution of multi-step ahead forecasts are approximated by Monte Carlo simulation. We show in experiments that our model produces accurate and sharp probabilistic forecasts. The estimated uncertainty of our forecasting also realistically increases over time, in a spontaneous manner.

Many models such as Long Short Term Memory (LSTMs), Gated Recurrent Units (GRUs) and transformers have been developed to classify time series data with the assumption that events in a sequence are ordered. On the other hand, fewer models have been developed for set based inputs, where order does not matter. There are several use cases where data is given as partially-ordered sequences because of the granularity or uncertainty of time stamps. We introduce a novel transformer based model for such prediction tasks, and benchmark against extensions of existing order invariant models. We also discuss how transition probabilities between events in a sequence can be used to improve model performance. We show that the transformer-based equal-time model outperforms extensions of existing set models on three data sets.

In many cases, it is difficult to generate highly accurate models for time series data using a known parametric model structure. In response, an increasing body of research focuses on using neural networks to model time series approximately. A common assumption in training neural networks on time series is that the errors at different time steps are uncorrelated. However, due to the temporality of the data, errors are actually autocorrelated in many cases, which makes such maximum likelihood estimation inaccurate. In this paper, we propose to learn the autocorrelation coefficient jointly with the model parameters in order to adjust for autocorrelated errors. For time series regression, large-scale experiments indicate that our method outperforms the Prais-Winsten method, especially when the autocorrelation is strong. Furthermore, we broaden our method to time series forecasting and apply it with various state-of-the-art models. Results across a wide range of real-world datasets show that our method enhances performance in almost all cases.

The majority of Chinese characters are monophonic, i.e.their pronunciations are unique and thus can be induced easily using a check table. As for their counterparts, polyphonic characters have more than one pronunciation. To perform linguistic computation tasks related to spoken Mandarin Chinese, the correct pronunciation for each polyphone must be identified among several candidates according to its context. This process is called Polyphone Disambiguation, a key procedure in the Grapheme-to-phoneme (G2P) conversion step of a Chinese text-to-speech (TTS) system. The problem is well explored with both knowledge-based and learning-based approaches, yet it remains challenging due to the lack of publicly available datasets and complex language phenomenon concerned polyphone. In this paper, we propose a novel semi-supervised learning (SSL) framework for Mandarin Chinese polyphone disambiguation that can potentially leverage unlimited unlabeled text data. We explore the effect of various proxy labeling strategies including entropy-thresholding and lexicon-based labeling. As for the architecture, a pre-trained model of Electra is combined with Convolution BLSTM layers to fine-tune on our task. Qualitative and quantitative experiments demonstrate that our method achieves state-of-the-art performance in Mandarin Chinese polyphone disambiguation. In addition, we publish a novel dataset specifically for the polyphone disambiguation task to promote further researches.

Both traditional automakers and Internet companies have long been involved in the development of automated driving technology and have achieved certain results. In 2017, GM equipped the Super Cruise automatic driving function on the Cadillac CT6. In April of the same year, Baidu released the Apollo self-driving vehicles platform. In July, Audi officially released the Audi A8, and its automated driving system Traffic Jam Pilot reached Level 3. In October, Waymo completed the first social road test of Level 4 self-driving vehicles for the first time. In April 2018, Baidu launched the test ride of Level 4 Baidu driverless bus "Apolon," and announced the automated driving bus entered the mass production phase in July. The rapid development of automated driving technology has also led to a lot of discussions - most of which are concerned about the widespread use of automated driving technology.