CS 294-131: Special Topics in Deep Learning

Instructors

Trevor Darrell

Dawn Song

Teaching Assistants

Robert Nishihara

Office Hours

Robert Nishihara: By appointment

Lectures

Time: Monday 1 - 2:30pm

Location: Soda 306

Piazza

Course announcements will be announced through Piazza. If you are in the class, sign up on Piazza.

For more information about deep learning at Berkeley, sign up for the talk announcement mailing list.

Syllabus

Poster Session

Date	Speaker	Readings	Talk	Deadlines
01/22	Stefano Soatto	Main Reading: Emergence of Invariance and Disentanglement in Deep Representations Stochastic Gradient Descent Performs Variational Inference, Converges to Limit Cycles for Deep Networks Background Reading: Information Bottleneck Method	The Emergence Theory of Deep Learning
01/29	Alison Gopnik	Main Reading: Changes in cognitive flexibility and hypothesis search across human life history from childhood to adolescence to adulthood Reconstructing constructivism: Causal models, Bayesian learning mechanisms and the theory theory Background Reading: When Younger Learners Can Be Better (or at Least More Open-Minded) Than Older Ones Probabilistic models, learning algorithms, and response variability: sampling in cognitive development	What 4 year olds can do and AI can’t (yet)
02/05	Mike Lewis	Main Reading: Deal or No Deal? End-to-End Learning for Negotiation Dialogues Hierarchical Text Generation and Planning for Strategic Dialogue Background Reading: Learning Cooperative Visual Dialog Agents with Deep Reinforcement Learning	Deal or No Deal? End-to-End Learning for Negotiation Dialogues
02/12	Kevin Murphy	Main Reading: Improved Image Captioning via Policy Gradient optimization of SPIDEr Generative Models of Visually Grounded Imagination Background Reading: Self-critical Sequence Training for Image Captioning SCAN: Learning Abstract Hierarchical Compositional Visual Concepts	Probabilistic models for vision and language	Final Project Proposal Due
02/19	President's Day
02/26	Thomas Funkhouser	Main Reading: 3DMatch: Learning Local Geometric Descriptors from RGB-D Reconstructions Interactive 3D Modeling with a Generative Adversarial Network Background Reading: Semantic Scene Completion from a Single Depth Image Learning a Probabilistic Latent Space of Object Shapes via 3D Generative-Adversarial Modeling	Data-Driven Methods for Matching, Labeling, and Synthesizing 3D Shapes
03/05	Dileep George	Main Reading: A generative vision model that trains with high data efficiency and breaks text-based CAPTCHAs Dynamic Routing Between Capsules Background Reading: Neural networks and the bias variance dilemma Hierarchical Bayesian inference in the visual cortex	Visual Perception, Data Efficiency, and Deep Learning
03/12	Bruno Olshausen	Main Reading: Emergence of foveal image sampling from learning to attend in visual scenes Building High-level Features Using Large Scale Unsupervised Learning Background Reading: Sparse Coding via Thresholding and Local Competition in Neural Circuits	Perception in Brains and Machines
03/19	Alex Kurakin	Main Reading: Adversarial Examples that Fool both Human and Computer Vision Adversarial Logit Pairing Background Reading: Delving into Transferable Adversarial Examples and Black-box Attacks Towards Deep Learning Models Resistant to Adversarial Attacks	Adversarial Examples	Final Project Milestone Due
03/26	Spring Break
04/02	Mohammad Norouzi	Main Reading: Deep Value Networks Learn to Evaluate and Iteratively Refine Structured Outputs Discovery of Semantic 3D Keypoints via End-to-end Geometric Reasoning Background Reading: Large Margin Methods for Structured and Interdependent Output Variables Transforming Autoencoders	Expressive Structured Models of Images and Objects
04/09	Abhinav Gupta	Main Reading: Transitive Invariance for Self-supervised Visual Representation Learning Learning by Asking Questions Background Reading: Unsupervised Visual Representation Learning by Context Prediction Unsupervised Learning of Visual Representations using Videos	SuperSizing and Empowering Visual Learning
04/16	Ryan Adams	Main Reading: Composing graphical models with neural networks for structured representations and fast inference Auto-Encoding Variational Bayes Background Reading: Stochastic Variational Inference Mapping Sub-Second Structure in Mouse Behavior	Building Probabilistic Structure into Massively Parameterized Models
04/23	POSTER SESSION	The poster session is 1-4pm in the Soda Hall 5th floor atrium. You can begin setting up at 12:30pm.
04/27				FINAL REPORT DUE

Course description

In recent years, deep learning has enabled huge progress in many domains including computer vision, speech, NLP, and robotics. It has become the leading solution for many tasks, from winning the ImageNet competition to winning at Go against a world champion. This class is designed to help students develop a deeper understanding of deep learning and explore new research directions and applications of deep learning. It assumes that students already have a basic understanding of deep learning. In particular, we will explore a selected list of new, cutting-edge topics in deep learning, including new techniques and architectures in deep learning, security and privacy issues in deep learning, recent advances in the theoretical and systems aspects of deep learning, and new application domains of deep learning such as autonomous driving.

Class format and project

This is a lecture, discussion, and project oriented class. Each lecture will focus on one of the topics, including a survey of the state-of-the-art in the area and an in-depth discussion of the topic. Each week, students are expected to complete reading assignments before class and participate actively in class discussion.

Students will also form project groups (two to three people per group) and complete a research-quality class project.

Enrollment information

For undergraduates: Please note that this is a graduate-level class. However, with instructors’ permission, we do allow qualified undergraduate students to be in the class. If you are an undergraduate student and would like to enroll in the class, please fill out this form and come to the first lecture of the class. Qualified undergraduates will be given instructor codes to be allowed to register for the class after the first lecture of the class, subject to space availability.

If you have not received grades for some classes that you are currently enrolled in, please choose Currently Enrolled and then update the form when you receive your final grades. You may also be interested in this class, which is open to undergraduates.

Students may enroll in this class for variable units.

1 unit: Participate in reading assignments (including serving as discussion lead once).
3 units: Both reading assignments and a project. Projects may fall into one of four categories:
- Traditional Literature Review of a deep learning topic (e.g., literature review of deep dialogue systems)
- Distill-like Literature Review of a deep learning topic (e.g., a Distill-like blog post illustrating different optimization techniques used in deep learning)
- Reimplement research code and open source it
- Conference level research project
You may not take this class for 4 units.

Deadlines

Reading assignment deadlines:
- For students,
  - Submit questions by Friday noon
  - Vote on the poll of discussion questions by Saturday 11:59 pm
- For discussion leads,
  - Send form to collect questions from students by Wednesday 11:59 pm
  - Summarize questions proposed by students to form the poll and send it by Friday 11:59 pm
  - Summarize the poll to generate a ranked & categorized discussion question list and send the list to teaching staff by Sunday 7pm
  - Answer all Piazza questions about the assigned readings, both the week before and the week after the lecture.

Grading

20% class participation
25% weekly reading assignment
- 10% discussion leads
- 15% individual reading assignments
55% project

Additional Notes

For students who need computing resources for the class project, we recommend you to look into AWS educate program for students. You’ll get 100 dollar’s worth of sign up credit. Here’s the link .