EE380 Computer Systems Colloquium

EE380 Computer Systems Colloquium presents "Instruction Sets Should Be Free: The Case For RISC-V"

Topic: 
Instruction Sets Should Be Free: The Case For RISC-V
Abstract / Description: 

The increasing popularity today of systems on a chip, where processors are just part of the design, calls into question why one of the most important interfaces is proprietary. We argue that:
There is no good technical reason not to have free, open instruction sets just as we have free, open networking standards and free, open operating systems.
The most likely first targets for a free, open instruction set are systems on a chip for the Internet of Things, which have low cost and power demands, and for Warehouse Scale Computers, which could benefit from viable alternatives to the 80x86 instruction set
The best architectural style for a free, open instruction set is RISC.


Given the time it takes to design an instruction set, it makes more sense to adopt an existing RISC free, open instruction set than to design a new one from scratch.
Among the existing RISC free, open instruction sets, RISC-V is the best and safest choice.

 

To learn more, see www.riscv.org

Date and Time: 
Wednesday, April 10, 2019 - 4:30pm
Venue: 
Shriram 104

EE380 Computer Systems Colloquium presents "Saving energy and increasing density in information processing using photonics"

Topic: 
Saving energy and increasing density in information processing using photonics
Abstract / Description: 

Information processing is increasingly limited by the energy required for interconnections and by the need for greater density of communications. Unfortunately, wires cannot scale to solve these problems. Optics and photonics allow us both to increase density and to reduce energy for communications on and off electronic chips and for all longer distances [1], they offer orders of magnitude of "headroom" for improving both density and energy, and they may be the only viable approach. Photonic technology has its challenges, but recent advances and future possibilities show that integrated photonics could provide the necessary technical solutions; indeed, such advanced photonics technology will be essential if our use of information is to continue to increase at current rates [1]. This talk will summarize the arguments for why we need such photonics and also some of the surprising directions we will need to consider; for example, for short distance interconnects inside and possibly between machines, to reduce energy we may need to avoid any time-multiplexing and exploit parallel free-space optics instead. Fortunately, technologies like silicon photonics could help us do this, though we need new generations of integration and optical system technology.

[1] D. A. B. Miller, Attojoule Optoelectronics for Low-Energy Information Processing and Communications: a Tutorial Review, IEEE/OSA J. Lightwave Technology 35, 343-393 (2017) DOI: 10.1109/JLT.2017.2647779

Date and Time: 
Wednesday, April 3, 2019 - 4:30pm
Venue: 
Shriram 104

EE380 Computer Systems Colloquium presents Karthik and Arushi

Topic: 
User Interface
Abstract / Description: 

Conversational applications often are over-hyped and under perform. While there's been significant progress in Natural Language Understanding (NLU) in academia and a huge growing market for voice based technologies, NLU performance significantly drops when you introduce language with typos or other errors, uncommon vocabulary, and more complex requests. This talk will cover how to build a production quality conversational app that performs well in a real world setting.

We will demonstrate an end-to-end approach for consistently building conversational interfaces with production-level accuracies that has proven to work well for a number of applications across diverse verticals. Building successful conversational interfaces involves choosing the right use case, collecting clean and relevant data, and breaking down the NLU problem into a series of solvable sub-tasks. All of today's most widely used conversational services have been built using a similar hierarchical NLU pipeline of domain-intent-entity classification that has become an industry standard, which we will discuss in detail.

Our architecture further improves on this standard domain-intent-entity classification and dialogue management architecture by leveraging shallow semantic parsing. We observed that NLU systems for industry applications often require more structured representations of entity relations than provided by the standard hierarchy, yet without requiring full semantic or syntactic parses which are often inaccurate on real-world conversational data. We describe our approach and demonstrate how it improves the performance of conversational interfaces for non-trivial use cases.

We end the talk by discussing the additional challenges in building a voice assistant rather than a text-based chatbot. Large vocabulary domain-agnostic Automatic Speech Recognition (ASR) systems often mis-transcribe domain-specific words and phrases. Since these generic ASR systems are the first components of most voice assistants in production, building NLU systems that are robust to these errors can be a challenging task. We describe a few potential methods for handling ASR errors in the NLU pipeline, especially in the entity classification and resolution component which is most susceptible to poor performance from ASR errors.

After this talk, attendees will have a better appreciation for the challenges and nuances of building real-world NLU systems, as well as a high level understanding of the best practices and components needed to build their own production quality conversational assistant.


The Stanford EE Computer Systems Colloquium (EE380) meets on Wednesdays 4:30-5:45 throughout the academic year. Talks are given before a live audience in Room 104 of the Shriram Building on the Stanford Campus. The live talks (and the videos hosted at Stanford and on YouTube) are open to the public.

Stanford students may enroll in EE380 to take the Colloquium as a one unit S/NC class. Enrolled students are required to keep and electronic notebook or journal and to write a short, pithy comment about each of the ten lectures and a short free form evaluation of the class in order to receive credit. Assignments are due at the end of the quarter, on the last day of examinations.

EE380 is a video class. Live attendance is encouraged but not required. We (the organizers) feel that watching the video is not a substitute for being present in the classroom. Questions are encouraged.

Many past EE380 talks are available on YouTube, see the EE380 Playlist.

Date and Time: 
Wednesday, March 13, 2019 - 4:30pm
Venue: 
Shriram 104

EE380 Computer Systems Colloquium presents Jamie Morgenstern

Topic: 
Extending the theory of ML for human-centric applications
Abstract / Description: 

Many recent application domains for machine learning deviate from standard modeling assumptions, by including data generated by people who may want to manipulate a system's output, or by trying to accomplish some task for which multiple objectives are simultaneously important. For example, an employer might want to promote their job opportunities to people with certain skills, while simultaneously ensuring a broad range of demographics sees and applies to the job posting. Moreover, if the employer uses a fixed filter to sift out fraudulent applications, the filter will become less useful over time as both fraudulent and honest applicants shift their application contents to pass the filter. In this talk, I will survey some recent results that take steps towards making ML methods more robust to these natural environments often faced in the real world.

Video: To access the live webcast of the talk (active at 16:28 of the day of the presentaton) and the archived version of the talk, use the URL SU-EE380-20190306. This is a first class reference and can be transmitted by email, Twitter, etc.

A URL referencing a YouTube view of the lecture will be posted a week or so following the presentation.


The Stanford EE Computer Systems Colloquium (EE380) meets on Wednesdays 4:30-5:45 throughout the academic year. Talks are given before a live audience in Room 104 of the Shriram Building on the Stanford Campus. The live talks (and the videos hosted at Stanford and on YouTube) are open to the public.

Stanford students may enroll in EE380 to take the Colloquium as a one unit S/NC class. Enrolled students are required to keep and electronic notebook or journal and to write a short, pithy comment about each of the ten lectures and a short free form evaluation of the class in order to receive credit. Assignments are due at the end of the quarter, on the last day of examinations.

EE380 is a video class. Live attendance is encouraged but not required. We (the organizers) feel that watching the video is not a substitute for being present in the classroom. Questions are encouraged.

Many past EE380 talks are available on YouTube, see the EE380 Playlist.

Date and Time: 
Wednesday, March 6, 2019 - 4:30pm
Venue: 
Shriram 104

EE380 Computer Systems Colloquium presents Judith Estrin

Topic: 
TBA
Abstract / Description: 

TBA


 

The Stanford EE Computer Systems Colloquium (EE380) meets on Wednesdays 4:30-5:45 throughout the academic year. Talks are given before a live audience in Room 104 of the Shriram Building on the Stanford Campus. The live talks (and the videos hosted at Stanford and on YouTube) are open to the public.

Stanford students may enroll in EE380 to take the Colloquium as a one unit S/NC class. Enrolled students are required to keep and electronic notebook or journal and to write a short, pithy comment about each of the ten lectures and a short free form evaluation of the class in order to receive credit. Assignments are due at the end of the quarter, on the last day of examinations.

EE380 is a video class. Live attendance is encouraged but not required. We (the organizers) feel that watching the video is not a substitute for being present in the classroom. Questions are encouraged.

Many past EE380 talks are available on YouTube, see the EE380 Playlist.

Date and Time: 
Wednesday, February 27, 2019 - 4:30pm
Venue: 
Shriram 104

EE380 Computer Systems Colloquium presents Babble Labble

Topic: 
Babble Labble: Training Classifiers with Natural Language Explanations
Abstract / Description: 

Training accurate classifiers requires many labels, but each label provides only limited information (one bit for binary classification). In this work, we propose BabbleLabble, a framework for training classifiers in which an annotator provides a natural language explanation for each labeling decision. A semantic parser converts these explanations into programmatic labeling functions that generate noisy labels for an arbitrary amount of unlabeled data, which is used to train a classifier. On three relation extraction tasks, we find that users are able to train classifiers with comparable F1 scores from 5-100 times faster by providing explanations instead of just labels. Furthermore, given the inherent imperfection of labeling functions, we find that a simple rule-based semantic parser suffices.

The full paper can be found here: https://arxiv.org/abs/1805.03818.


The Stanford EE Computer Systems Colloquium (EE380) meets on Wednesdays 4:30-5:45 throughout the academic year. Talks are given before a live audience in Room 104 of the Shriram Building on the Stanford Campus. The live talks (and the videos hosted at Stanford and on YouTube) are open to the public.

Stanford students may enroll in EE380 to take the Colloquium as a one unit S/NC class. Enrolled students are required to keep and electronic notebook or journal and to write a short, pithy comment about each of the ten lectures and a short free form evaluation of the class in order to receive credit. Assignments are due at the end of the quarter, on the last day of examinations.

EE380 is a video class. Live attendance is encouraged but not required. We (the organizers) feel that watching the video is not a substitute for being present in the classroom. Questions are encouraged.

Many past EE380 talks are available on YouTube, see the EE380 Playlist.

Date and Time: 
Wednesday, February 20, 2019 - 4:30pm
Venue: 
Shriram 104

EE380 Computer Systems Colloquium presents Electronic Design Automation (EDA) and the Resurgence of Chip Design

Topic: 
Electronic Design Automation (EDA) and the Resurgence of Chip Design
Abstract / Description: 

Electronic Design Automation (EDA) enables the design of semiconductors comprised of tens of billions of devices. EDA consists of design software incorporating cutting-edge optimization and analysis algorithms, as well as pre-designed blocks known as "Intellectual Property" (IP). This talk reviews trends in the EDA industry in the context of the Semiconductor industry, highlighting new developments such as the rise of IP to become the largest segment in EDA, the increased use of system-level tools, and the resurgence of chip design particularly in the area of artificial intelligence. The talk ends with a brief mention of two related activities: The Stanford class "EDA and Machine Learning Hardware", in which students learn the inner workings of (some) EDA tools and how to use them effectively to design digital hardware, e.g. a convolutional neural network for image recognition implemented in an FPGA; and Silicon Catalyst, an innovative local incubator focused on accelerating solutions in Silicon.

Date and Time: 
Wednesday, February 6, 2019 - 4:30pm
Venue: 
Shriram 104

EE380 Computer Systems Colloquium presents "Scalable Intelligent Systems Build and Deploy by 2025"

Topic: 
Scalable Intelligent Systems Build and Deploy by 2025
Abstract / Description: 

The next stage of human-computer evolution, Scalable Intelligent Systems, integrates people, communications, and computers into a unified cooperative environment. Think of it as moving beyond Google, Facebook, Instagram, and the other social networks. Scalable Intelligent Systems can be actualized by 2025 and are likely to include the following characteristics:

  • Interactively acquire and present information from video, web pages, hologlasses, online data bases, sensors, articles, human speech and gestures, etc.
  • Real-time integration of massive, pervasively inconsistent information
  • Close human collaboration using hologlasses for secure mobile interaction.
  • Organizations of people and IoT devices (Citadels) for trustworthiness, resilience, and performance with no single point of failure
  • Scalability in all important dimensions including no hard barriers to continual improvement in the above areas.

There is no computer-only solution that can implement the above by 2025. Consequently, people are fundamental to a Scalable Intelligent System.

Scalable Intelligent Systems, as envisioned, will be the most complex software that has ever been created. Every advanced country in the world has recently introduced its own development plan. The development of Scalable Intelligent Systems will create enormous social and policy challenges.

For example, Scalable Intelligent Systems can be of enormous value in Pain Management. Pain management requires much more than just prescribing opioids and other pain killers, which are often critical for short-term and less often longer-term use. Organizational aspects play an important role in pain management. Scalable Intelligent Systems can help users with appliances, entertainment, exercise, hypnosis, medication, meditation, physical therapy, and collaboration with medical helpers.

Building and deploying a Scalable Intelligent Technology stack is possible by 2025. The rewards are high but the amount of effort required is substantial. We have to get to work.

 

Date and Time: 
Wednesday, January 23, 2019 - 4:30pm
Venue: 
Shriram 104

EE380 Computer Systems Colloquium presents Erudite: A Low-Latency, High-Capacity, and High-efficiency Prototype System for Computational Intelligence

Topic: 
Erudite: A Low-Latency, High-Capacity, and High-efficiency Prototype System for Computational Intelligence
Abstract / Description: 

Since the rise of deep learning in 2012, much progress has been made in deep-learning-based AI tasks such as image/video understanding and natural language understanding, as well as GPU/accelerator architectures that greatly improve the training and inference speed for neural-network models. As the industry players race to develop ambitious applications such as self-driving vehicles, cashier-less supermarkets, human-level interactive robot systems, and human intelligence augmentation, major research challenges remain in computational methods as well as hardware/software infrastructures required for these applications to be effective, robust, responsive, accountable and cost-effective. Innovations in scalable iterative solvers and graph algorithms will be needed to achieve these application-level goals but will also impose much higher-level of data storage capacity, access latency, energy efficiency, and processing throughput. In this talk, I will present our recent progress in building highly performant AI task libraries, creating full AI applications, providing AI application development tools, and prototyping the Erudite system at the IBM-Illinois C3SR.

Date and Time: 
Wednesday, January 16, 2019 - 4:30pm
Venue: 
Shriram 104

John G. Linvill Distinguished Seminar on Electronic Systems Technology

Topic: 
Internet of Things and Internet of Energy for Connecting at Any Time and Any Place
Abstract / Description: 

In this presentation, I would like to discuss with you how to establish a sustainable and smart society through the internet of energy for connecting at any time and any place. I suspect that you have heard the phrase, "Internet of Energy" less often. The meaning of this phrase is simple. Because of a ubiquitous energy transmission system, you do not need to worry about a shortage of electric power. One of the most important items for establishing a sustainable society is [...]


"Inaugural Linvill Distinguished Seminar on Electronic Systems Technology," EE News, July 2018

 

Date and Time: 
Monday, January 14, 2019 - 4:30pm
Venue: 
Hewlett 200

Pages

Subscribe to RSS - EE380 Computer Systems Colloquium