Goto

Collaborating Authors

 Technology


Unsupervised Domain Adaptation with Residual Transfer Networks

Neural Information Processing Systems

The recent success of deep neural networks relies on massive amounts of labeled data. For a target task where labeled data is unavailable, domain adaptation can transfer a learner from a different source domain. In this paper, we propose a new approach to domain adaptation in deep networks that can jointly learn adaptive classifiers and transferable features from labeled data in the source domain and unlabeled data in the target domain. We relax a shared-classifier assumption made by previous methods and assume that the source classifier and target classifier differ by a residual function. We enable classifier adaptation by plugging several layers into deep network to explicitly learn the residual function with reference to the target classifier. We fuse features of multiple layers with tensor product and embed them into reproducing kernel Hilbert spaces to match distributions for feature adaptation. The adaptation can be achieved in most feed-forward models by extending them with new residual layers and loss functions, which can be trained efficiently via back-propagation. Empirical evidence shows that the new approach outperforms state of the art methods on standard domain adaptation benchmarks.


Toward Deeper Understanding of Neural Networks: The Power of Initialization and a Dual View on Expressivity

Neural Information Processing Systems

We develop a general duality between neural networks and compositional kernel Hilbert spaces. We introduce the notion of a computation skeleton, an acyclic graph that succinctly describes both a family of neural networks and a kernel space. Random neural networks are generated from a skeleton through node replication followed by sampling from a normal distribution to assign weights. The kernel space consists of functions that arise by compositions, averaging, and non-linear transformations governed by the skeleton's graph topology and activation functions. We prove that random networks induce representations which approximate the kernel space. In particular, it follows that random weight initialization often yields a favorable starting point for optimization despite the worst-case intractability of training neural networks.



'I'll key your car': ChatGPT can become abusive when fed real-life arguments, study finds

The Guardian

In some cases, ChatGPT's outputs went beyond those of the human participants, including personalised insults and explicit threats. In some cases, ChatGPT's outputs went beyond those of the human participants, including personalised insults and explicit threats. 'I'll key your car': ChatGPT can become abusive when fed real-life arguments, study finds ChatGPT can escalate into abusive and even threatening language when drawn into prolonged, human-style conflict, according to a new study. Researchers tested how large language models (LLMs) responded to sustained hostility by feeding ChatGPT exchanges from real-life arguments and tracking how its behaviour changed over time. One expert not connected with the study described it as "one of the most interesting ever done into AI language and pragmatics Dr Vittorio Tantucci, who co-authored the research paper with Prof Jonathan Culpeper at Lancaster University, said their research found AI mirrored the dynamics of real-world disputes.


Tim Cook's Legacy Is Turning Apple Into a Subscription

WIRED

Tim Cook's Legacy Is Turning Apple Into a Subscription The soon-to-exit Apple CEO went all in on services. Now, the incoming CEO, John Ternus, will need to embrace the AI era. Tim Cook's tenure as CEO at Apple, which is coming to a close September 1, will likely be defined by operational efficiency and financial growth, ushering Apple into its trillion-dollar era. But his most significant achievement might be in doubling down on Apple's services business, which includes iCloud, the App Store, Apple Music, Apple TV+, News+, and more. It's the subscription layer on top of iOS, and almost all of the service apps are tightly integrated with Messages, the glue that keeps people stuck to their iPhones .


How to tell eagle parents Jackie and Shadow apart

Popular Science

Jackie looks stern, while Shadow looks a bit more surprised. More information Adding us as a Preferred Source in Google by using this link indicates that you would like to see more of our content in Google News results. Jackie and Shadow have been nesting together since 2018. Breakthroughs, discoveries, and DIY tips sent six days a week. The two new eaglets eating, chirping, and "bopping" in their nest high above Southern California's Big Bear Lake are arguably the stars of the popular wildlife livestream .


How we discovered the speed limit of arithmetic โ€“ and broke it

New Scientist

Some seemingly simple sequences of multiplication and addition grow so quickly that they question the very foundations of mathematics. Did you hear the one about the man who invented chess and got himself executed? Legend has it that a man called Sessa, who lived in India long ago, developed the rules for the game and presented them to a king. The king was delighted and offered the man his pick of reward. Sessa asked for a supposedly humble quantity of rice.


Google now lets you have full conversations with Gemini for Home

Engadget

The feature is rolling out for all the smart home program's supported languages and regions. Google announced today that it is upgrading the Gemini for Home service with a continued conversations feature. Continued conversation allows a user to have a natural discussion with the Gemini platform without prefacing every follow-up request with the Hey Google prompt. The microphone will remain active on a smart device for a few seconds after the Gemini AI assistant provides its reply. During that window, the lights on the hardware will pulse or glow, indicating that you can keep chatting normally with the chatbot without needing a wake word.


Robot Talk Episode 148 โ€“ Ethical robot behaviour, with Alan Winfield

Robohub

Alan Winfield is Professor of Robot Ethics at the University of the West of England (UWE), Visiting Professor at the University of York, and Associate Fellow of the Cambridge Centre for the Future of Intelligence. Alan co-founded the Bristol Robotics Laboratory, where his research is focussed on the science, engineering and ethics of cognitive robotics. Alan is an advocate for robot ethics; he chairs the advisory board of the Responsible Technology Institute at the University of Oxford and has co-drafted new standards on ethical risk assessment and transparency. Robot Talk is a weekly podcast that explores the exciting world of robotics, artificial intelligence and autonomous machines. Robot Talk is a weekly podcast that explores the exciting world of robotics, artificial intelligence and autonomous machines.


Robot Talk Episode 150 โ€“ House building robots, with Vikas Enti

Robohub

Claire chatted to Vikas Enti from Reframe Systems about using robotics and automation to build climate-resilient, high-performance homes. Vikas Enti is the co-founder and CEO of Reframe Systems, a physical AI company rethinking how homes are built through automation and localized fabrication. He previously spent more than a decade at Amazon Robotics, where he helped scale advanced robotics systems across global logistics networks. Today, he is applying those same principles of systems design and repeatable production to address the housing shortage. Vikas focuses on building climate-resilient, high-performance homes faster and more predictably than traditional methods.