We present an empirical investigation of pre-trained Transformer-based auto-regressive language models for the task of open-domain dialogue generation. Training paradigm of pre-training and fine-tuning is employed to conduct the parameter learning. Corpora of News and Wikipedia in Chinese and English are collected for the pre-training stage respectively. Dialogue context and response are concatenated into a single sequence utilized as the input of the models during the fine-tuning stage. A weighted joint prediction paradigm for both context and response is designed to evaluate the performance of models with or without the loss term for context prediction. Various of decoding strategies such as greedy search, beam search, top-k sampling, etc. are employed to conduct the response text generation. Extensive experiments are conducted on the typical single-turn and multi-turn dialogue corpora such as Weibo, Douban, Reddit, DailyDialog, and Persona-Chat. Detailed numbers of automatic evaluation metrics on relevance and diversity of the generated results for the languages models as well as the baseline approaches are reported.

Planning is an essential topic in the realm of automated driving. Besides planning algorithms that are widely covered in the literature, planning requires different software tools for its development, validation, and execution. This paper presents a survey of such tools including map representations, communication, traffic rules, open-source planning stacks and middleware, simulation, and visualization tools as well as benchmarks. We start by defining the planning task and different supporting tools. Next, we provide a comprehensive review of state-of-the-art developments and analysis of relations among them. Finally, we discuss the current gaps and suggest future research directions.

We present new planning and learning algorithms for RAE, the Refinement Acting Engine. RAE uses hierarchical operational models to perform tasks in dynamically changing environments. Our planning procedure, UPOM, does a UCT-like search in the space of operational models in order to find a near-optimal method to use for the task and context at hand. Our learning strategies acquire, from online acting experiences and/or simulated planning results, a mapping from decision contexts to method instances as well as a heuristic function to guide UPOM. Our experimental results show that UPOM and our learning strategies significantly improve RAE's performance in four test domains using two different metrics: efficiency and success ratio.

We consider the problem of zero-shot coordination - constructing AI agents that can coordinate with novel partners they have not seen before (e.g. humans). Standard Multi-Agent Reinforcement Learning (MARL) methods typically focus on the self-play (SP) setting where agents construct strategies by playing the game with themselves repeatedly. Unfortunately, applying SP naively to the zero-shot coordination problem can produce agents that establish highly specialized conventions that do not carry over to novel partners they have not been trained with. We introduce a novel learning algorithm called other-play (OP), that enhances self-play by looking for more robust strategies, exploiting the presence of known symmetries in the underlying problem. We characterize OP theoretically as well as experimentally. We study the cooperative card game Hanabi and show that OP agents achieve higher scores when paired with independently trained agents. In preliminary results we also show that our OP agents obtains higher average scores when paired with human players, compared to state-of-the-art SP agents.

The human brain, fed on just the calorie input of a modest diet, easily outperforms state-of-the-art supercomputers powered by full-scale station energy inputs. The difference stems from the multiple states of brain processes versus the two binary states of digital processors, as well as the ability to store information without power consumption--non-volatile memory. These inefficiencies in today's conventional computers have prompted great interest in developing synthetic synapses for use in computers that can mimic the way the brain works. Now, researchers at King's College London, UK, report in ACS Nano Letters an array of nanorod devices that mimic the brain more closely than ever before. The devices may find applications in artificial neural networks.

The human brain, fed on just the calorie input of a modest diet, easily outperforms state-of-the-art supercomputers powered by full-scale station energy inputs. The difference stems from the multiple states of brain processes versus the two binary states of digital processors, as well as the ability to store information without power consumption--non-volatile memory. These inefficiencies in today's conventional computers have prompted great interest in developing synthetic synapses for use in computers that can mimic the way the brain works. Now, researchers at King's College London, UK, report in ACS Nano Letters an array of nanorod devices that mimic the brain more closely than ever before. The devices may find applications in artificial neural networks.

A few years ago, Amazon employed a new automated hiring tool to review the resumes of job applicants. Shortly after launch, the company realized that resumes for technical posts that included the word "women's" (such as "women's chess club captain"), or contained reference to women's colleges, were downgraded. The answer to why this was the case was down to the data used to teach Amazon's system. Based on 10 years of predominantly male resumes submitted to the company, the "new" automated system in fact perpetuated "old" situations, giving preferential scores to those applicants it was more "familiar" with. Defined by AI4ALL as the branch of computer science that allows computers to make predictions and decisions to solve problems, artificial intelligence (AI) has already made an impact on the world, from advances in medicine, to language translation apps.

Distributing malware by attaching tainted documents to emails is one of the oldest tricks in the book. It's not just a theoretical risk--real attackers use malicious documents to infect targets all the time. So on top of its anti-spam and anti-phishing efforts, Gmail expanded its malware detection capabilities at the end of last year to include more tailored document monitoring. At the RSA security conference in San Francisco on Tuesday, Google's security and anti-abuse research lead Elie Bursztein will present findings on how the new deep-learning scanner for documents is faring against the 300 billion attachments it has to process each week. It's challenging to tell the difference between legitimate documents in all their infinite variations and those that have specifically been manipulated to conceal something dangerous.

Perhaps this dog would prefer driving the car, just like three dogs that were trained to do so. We've all seen dogs traveling in cars, including how they like to peek out an open window and enjoy the fur-fluffing breeze and dwell in the cacophony of scents that blow along in the flavorful wind. Dogs have also frequently been used as living props in commercials for cars, pretending in some cases to drive a car, such as the Subaru "Barkleys" advertising campaign that initially launched on TV in 2018 and continued in 2019, proclaiming that Subaru cars were "officially" dog tested and dog approved. What you might not know or might not remember is that there were three dogs that were trained on driving a car and had their moment of unveiling in December of 2012 when they were showcased by driving a car on an outdoor track (the YouTube posted video has amassed millions of views). Yes, three dogs named Monty, Ginny, and Porter were destined to become the first true car drivers on behalf of the entire canine family. Monty at the time was an 18-month-old giant schnauzer cross, while the slightly younger Ginny at one year of age was a beardie whippet cross, and Porter was a youthful 10-month-old beardie.

Are Artificial Intelligence and Machine Learning really the right metaphors to address training sets that feed into automated processes? Kate Crawford and Trevor Paglen look into the production of training data and uncover the historical origins, labor practices, infrastructures, and epistemological assumptions, with biases and skews built into them from the outset. About SXSW: SXSW dedicates itself to helping creative people achieve their goals. Founded in 1987, SXSW is best known for its conference and festivals that celebrate the convergence of the interactive, film, and music industries. An essential destination for global professionals, the event features sessions, showcases, screenings, exhibitions, and a variety of networking opportunities.