Author Archives: Shota Momma

Allen in CICS, Thursday 11/19 at 12:00

Thursday, November 19, 2020 – 12:00 Kelsey Allen (MIT): Robotics

I am currently a PhD candidate under the supervision of Josh Tenenbaum in the Computational Cognitive Science group. Previously I was an intern at DeepMind, and received my B.Sc. from the University of British Columbia in Physics.

Thursday, November 19, 2020 – 12:00 Kelsey Allen (MIT): Robotics

Bio: I am currently a PhD candidate under the supervision of Josh Tenenbaum in the Computational Cognitive Science group. Previously I was an intern at DeepMind, and received my B.Sc. from the University of British Columbia in Physics.

About

The Machine Learning and Friends Lunch (MLFL) series is sponsored by our friends at Oracle Labs.

MLFL is a lively and interactive forum held weekly where friends of the UMass Amherst machine learning community can gather virtually and give or hear a 50-minute presentation on recent machine learning research.

What is it?   A gathering of students/faculty/staff with broad interest in the methods and applications of machine learning.
When is it?    Thursdays 12:00pm to 1:00pm, via Zoom
Who is invited?   Everyone is welcome.
More info? Email cds-info@cs.umass.edu with questions or suggestions.

Gaston in Cognitive Brownbag, Wednesday, Nov 18 at noon

The final cognitive brownbag is today and it’s Phoebe Gaston (UMD PhD, presently UConn). The talk will begin at 12:00, at the following Zoom room:

https://umass-amherst.zoom.us/j/97489074821?pwd=MG9Nbk9OS1g4N3RXMVZab3duQUtSQT09

Meeting ID: 974 8907 4821
Passcode: PBScog

The role of syntactic prediction in auditory word recognition
Sentence context is understood to play some role in the processing of bottom-up acoustic input during word recognition, but the nature of this influence is both unclear and controversial. In this talk, I will focus on the mechanism by which expectations for the syntactic category of an upcoming word are integrated with the auditory input that allows that word to be recognized. One possibility is that syntactic predictions could completely inhibit contextually inappropriate lexical candidates, but an alternative considered less often is that the category prediction instead facilitates items that match its constraints, without affecting items that don’t. I will argue that failure to account for how these two possibilities would manifest differently in dependent measures may help explain conflict in the literature on this question. I will then present our high-powered experiment in the visual world paradigm, specifically designed to distinguish between an inhibitory and a facilitatory mechanism for the category constraint. We found that wrong-category lexical candidates do demonstrate phonological competition, ruling out complete inhibition as the mechanism for the syntactic category constraint. Finally, turning to the nature of syntactic category predictions themselves as an important angle on the problem of the syntactic constraint, I will present a MEG study on the generation of syntactic predictions and the difficulty of disentangling lexical and syntactic prediction.

De-Arteaga in CICS, Thursday, November 12, at 11:00am

Thursday, November 12, 2020 – 11:00 Maria De-Arteaga (UT Austin): Human-Centered Machine Learning

Bio: I am an Assistant Professor at the Information, Risk and Operation Management Department at McCombs School of Business at the University of Texas at Austin. I am also a core faculty member in the interdepartmental Machine Learning Laboratory.

About

The Machine Learning and Friends Lunch (MLFL) series is sponsored by our friends at Oracle Labs.

MLFL is a lively and interactive forum held weekly where friends of the UMass Amherst machine learning community can gather virtually and give or hear a 50-minute presentation on recent machine learning research.

What is it?   A gathering of students/faculty/staff with broad interest in the methods and applications of machine learning.
When is it?    Thursdays 12:00pm to 1:00pm, via Zoom
Who is invited?   Everyone is welcome.
More info? Email cds-info@cs.umass.edu with questions or suggestions.

Parikh in CICS, Thursday 11/5 at 12:00

Ankur Parikh (Google) will give a talk in Machine Learning and Friends lunch talk series. The talk is entitled “Natural Language Processing.” Bio:

Ankur is a Research Scientist at Google NYC and his primary interests are in natural language processing and machine learning. Ankur received his PhD from Carnegie Mellon in 2015 (advised by Prof. Eric Xing) and his B.S.E. from Princeton University in 2009. He has received a best paper runner up award at EMNLP 2014 and a best paper in translational bioinformatics at ISMB 2011.

About

The Machine Learning and Friends Lunch (MLFL) series is sponsored by our friends at Oracle Labs.

MLFL is a lively and interactive forum held weekly where friends of the UMass Amherst machine learning community can gather virtually and give or hear a 50-minute presentation on recent machine learning research.

What is it?   A gathering of students/faculty/staff with broad interest in the methods and applications of machine learning.
When is it?    Thursdays 12:00pm to 1:00pm, via Zoom
Who is invited?   Everyone is welcome.
More info? Email cds-info@cs.umass.edu with questions or suggestions.

Bullard in CICS, Thursday 10/29 at noon

Thursday, October 29, 2020 – 12:00 Kalesha Bullard (Facebook Artificial Intelligence Research): Multi-Agent Reinforcement Learning

Bio: I am a Postdoctoral Researcher at Facebook Artificial Intelligence Research (FAIR), working in Multi-Agent Reinforcement Learning. In particular, I am currently exploring the problem space of multi-agent communication for embodied agents.

About

The Machine Learning and Friends Lunch (MLFL) series is sponsored by our friends at Oracle Labs.

MLFL is a lively and interactive forum held weekly where friends of the UMass Amherst machine learning community can gather virtually and give or hear a 50-minute presentation on recent machine learning research.

What is it?   A gathering of students/faculty/staff with broad interest in the methods and applications of machine learning.
When is it?    Thursdays 12:00pm to 1:00pm, via Zoom
Who is invited?   Everyone is welcome.
More info? Email cds-info@cs.umass.edu with questions or suggestions.